DroneXL: Drone News, DJI Rumors, Reviews, and Drone Laws

Oregon Warns Drone Pilots: One Flight Can Ground Every Fire Aircraft

Haye Kesteloo — Sun, 23 Aug 2026 14:24:00 GMT

Oregon’s two airspace agencies have a blunt message for drone pilots in the middle of the state’s worst wildfire season on record: if you fly, firefighters can’t. The Oregon Department of Aviation and the Oregon Department of Emergency Management issued a joint warning on August 20 telling operators to keep drones out of active fire zones entirely, after unauthorized flights delayed aerial firefighting inside the state.

More than 20 large fires are burning in Oregon, with roughly 2.4 million acres (about 9,700 square kilometers) already scorched, more than the state lost in all of 2025. When a single unauthorized drone shows up over a fire, planes and helicopters leave the airspace until it is confirmed clear, and those lost minutes let the fire grow.

DroneXL has reported on drones grounding Oregon’s firefighting aircraft since 2020. What has changed is the volume.

Oregon’s Warning Follows Drone Delays at the Grasshopper Fire

The Oregon Department of Aviation and the Oregon Department of Emergency Management published their joint warning on August 20, after drones delayed aerial operations at the Grasshopper Fire near Mount Hood. The agencies’ message to pilots is blunt: if you fly, firefighting aircraft can’t.

“We want to make sure that people understand: if there’s a wildfire, don’t fly,” Oregon Department of Aviation Director Kenji Sugahara told KOIN 6 News. Sugahara said pilots flying waterdrops move at high speed and low altitude through smoke they cannot see through, and that crews had been hearing about drone problems around the Grasshopper Fire that were delaying operations.

Andrew Jahier, the emergency management department’s director of response and recovery, put a price on it in the release. “No photo is worth a federal prison sentence or a $75,000 fine,” Jahier said.

Firefighting aircraft work the same couple hundred feet of air that consumer drones occupy. The release describes low-altitude tankers and helicopters sharing airspace with recreational drones, a collision risk that ends the mission the moment a drone appears.

Drone Incursions Have Grounded Aircraft Across the West This Month

Oregon is not an outlier. The US Forest Service has reported 67 drone incursions near wildfires nationally so far in 2026, and August has been the worst stretch: grounded aircraft over the Spokane Complex in Washington and repeated shutdowns at New Mexico’s Frijoles Fire.

The Spokane numbers are the ugliest. Spokane County Sheriff John Nowels reported 26 drone incursions in a single day on August 2, four of them directly in the path of working aircraft. The next evening a drone grounded the fleet again for about 30 minutes. The sheriff’s office has since identified multiple operators and referred citations to the Federal Aviation Administration.

In New Mexico, drones grounded air tankers and helicopters twice in one week over the Frijoles Fire in the Santa Fe National Forest.

This follows a record 2025. The Forest Service counted 218 drone incursions over wildfires last year, 184 of them at the Eaton and Palisades Fires alone, more than the 125 incidents the agency logged across the previous seven years combined. That season produced the collision between a DJI Mini and a Super Scooper over the Palisades Fire, and the operator pleaded guilty to a federal charge.

A drone strike left a 3-inch by 6-inch hole in the left wing of a Super Scooper firefighting aircraft battling the Palisades Fire near Malibu, California. (FBI photo)

Oregon has its own history here. A drone grounded firefighting aircraft over an Oregon wildfire in June 2024, and the state’s newest warning reads almost word for word like the ones that came before it.

Violators Face $75,000 Fines and Federal Prison Time

Flying a drone over an Oregon wildfire can cost up to $75,000 in civil fines and up to 12 months in federal prison. Violators also face confiscation of their equipment and permanent loss of their pilot certificates. Detection gear deployed near fire zones is helping the state find them.

Sugahara told KOIN 6 News that drone-detection equipment at some incidents can locate a drone in the air and, in some cases, the ground station it launched from. (Spokane County identified multiple operators within days.) Anonymous fire footage is getting harder to pull off, and I won’t pretend to be sorry about that.

The agencies want every operator checking Temporary Flight Restrictions (TFRs) before takeoff. Active wildfire areas are no-drone zones around the clock, at night and on weekends, and anyone who spots an unauthorized drone in a fire zone should report it to local law enforcement or call 911.

DroneXL’s Take

Oregon’s warning deserves a wider audience than a state press release will get, because the operators flying over these fires are burning through goodwill the whole hobby depends on. DroneXL first covered illegal drones grounding aircraft over Oregon wildfires back in September 2020, and the story has barely changed.

Don’t be that guy. Nobody’s footage is worth a tanker crew pulling off a fire while houses burn below. Sixty-seven incursions this year. Every one of them is an argument handed to somebody drafting drone restrictions, and those drafts are already circulating: Grant County’s administrator is pushing a Washington state law creating criminal penalties for wildfire drone flights, drafted directly in response to Spokane.

Congress is further along. Oregon Congresswoman Janelle Bynum’s Wildfire Aerial Response Safety Act, H.R. 6618, passed the House on March 24 and now sits in the Senate. It only orders the FAA to study drone incursions, but lawmakers have already floated counter-drone powers at wildfires, and a fire season like this one is exactly what moves a study bill toward something with teeth. Watch whether the Senate takes it up before the season ends. My bet is that if the incursion count keeps climbing the way it has this August, the next bill written won’t stop at a study.

Sources: Oregon Department of Emergency Management, KOIN 6 News, FireRescue1, Albuquerque Journal

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

FCC Labels Drone Light Shows Military-Grade While US Operators Fly Korean And Chinese Fleets

Haye Kesteloo — Sat, 22 Aug 2026 20:45:00 GMT

The FCC wrote drone light shows into its definition of military-grade equipment by name. Category 7 of Public Notice DA 26-758, the July proposal to cut off imports and sales of previously authorized foreign drones, covers swarming systems, and its second bullet reaches aircraft built for synchronized formations, “including applications such as multi-UAS light shows.” Comments in PS Docket 26-189 close September 2.

The industry that sentence describes runs overwhelmingly on foreign-built aircraft. The largest American operators fly Korean-made fleets, Chinese manufacturers supply most of the world’s show drone equipment, and even the company Kimbal Musk built from Intel’s old light show division runs a factory in China.

I have photographed these shows myself, from 600 drones rising over the IJ in Amsterdam in January 2023 to the 1,000-drone Franchise Freedom performance over Central Park that fall. The aircraft I watched that night in New York now sit inside a national security definition.

The FCC Puts Light Shows Inside Its Swarm Category

Category 7 defines swarming UAS two ways: autonomous systems that coordinate without human intervention, and aircraft purpose-built for synchronized formation flight. The second definition names multi-drone light shows as an example. Six other categories cover thermal imaging, LiDAR, docking stations, sprayers, defense-article aircraft, and anything at 55 pounds or more.

We broke down the full seven-category proposal when it landed on July 21. The mechanics matter here. The Federal Register published the notice on August 3, setting the September 2 comment deadline, and the proposal carries a 180-day wind-down after any final action. Marketing, in FCC usage, covers advertising, distribution, sale, and lease. A foreign-built show drone on the Covered List could no longer be imported or sold, and covered replacement components would dry up with it.

Existing fleets keep flying. The proposal exempts continued use of equipment already purchased, along with Blue UAS listed products, Buy American qualifying hardware, and anything holding a Conditional Approval from the Department of War or Homeland Security. The squeeze lands on fleet growth and replacement, which is where a show business lives. Operators add aircraft every season, and a crashed or aging drone needs a successor from somewhere.

The December 2025 Covered List action already blocked new authorizations for foreign show drones. As we reported last week, the July proposal goes further by reaching hardware the FCC already approved.

American Show Operators Fly Korean-Built Aircraft

Most of the American drone light show industry flies one aircraft: the UVify IFO, built in South Korea. UVify’s chief operating officer Robert Cheek put the number at roughly 70 percent of American show companies in January, with about 80 percent of the global light show market on the company’s hardware.

UVify is headquartered in Seattle with a sales office in San Francisco, and it raised nearly $42 million in a Series B round this February. None of that changes the analysis under the Covered List entry, which turns on where the aircraft is produced. A Korean-built IFO is foreign-produced covered equipment, and a formation-flight aircraft is the textbook Category 7 machine.

Sky Elements, the Texas company that bills itself as the largest show provider in the country, flies IFO fleets. Brightflight Drone Shows in Denver has grown past 800 aircraft at roughly $1,000 each while flying the Colorado 150 tour we covered in January, and it cleared a fresh FAA altitude waiver this week authorizing 500-foot operations over Highlands Ranch. That is the regulatory posture of this industry: FAA waivers, visual observers, NOTAMs, geofences. The FCC now files the aircraft category under military capability.

Domestic builders exist. Firefly Drone Shows in Detroit engineers its aircraft in-house and markets itself as the only US-built show operator. Verge Aero, whose X-series hardware we covered back in 2022, designs and builds its own aircraft. Both stand to gain from a proposal that removes their foreign competition, and neither comes close to supplying the whole industry today.

Chinese Manufacturers Supply Most Of The World’s Show Drones

On the equipment side, the global market belongs to Shenzhen. A QYResearch report that Damoda promotes puts its share of drone show equipment at 36.42 percent worldwide. Crosstown rival HighGreat holds 24.12 percent, with Chinese suppliers near 80 percent combined.

Those are figures a market leader circulates about itself, and they track with everything visible in the field. Damoda holds six Guinness records, flew 10,197 drones from a single computer in September 2024, and made its North American debut over San Diego in May 2025 with a 1,000-drone show. The company flew CES 2026 in Las Vegas this past January as its listed official drone show operator. HighGreat claims annual production capacity of one million drones from its Longgang facility.

The dual-use question around this hardware is not imaginary. When Damoda unveiled a containerized system last fall that auto-deploys 648 drones from a truck at one click, The War Zone covered it as a signal of evolving swarm threats, while noting the system is built for entertainment and flies pre-scripted, localized routines.

Nova Sky Stories Manufactures In The US, Europe, And China

The strangest supply chain in the industry belongs to the company with the most American story. Nova Sky Stories is what became of Intel’s drone light show division, which Kimbal Musk bought in 2022 along with 9,000 Shooting Star aircraft.

Those Intel-lineage drones flew Super Bowl halftime shows, the Tokyo Olympics, and the Central Park performance we covered in October 2023. The 600 drones I photographed over Amsterdam at the A’DAM Tower show in January 2023 were Nova aircraft from that same Intel bloodline.

Preparations on the IJ waterfront ahead of the January 2023 drone light show in Amsterdam: crews wire up the launch grid, and Stephen Sutton of the Fly By Guys and a Nova Sky Stories crew member hold the Intel Shooting Star drones on the floating launch platform. Photo credit: Haye Kesteloo / DroneXL

Boulder-based Nova raised $50 million last August, with Jeffrey Katzenberg joining the board through WndrCo. Asked about tariffs on Yahoo Finance, Musk said the company builds its own drones in factories in the US, Europe, and China: “You’ve got to be able to manufacture in the markets you’re going to serve.” Under a country-of-production rule, the China-built portion of an American founder’s fleet sits on the Covered List next to Damoda’s.

Creative Skies Tells Show Operators To File Before September 2

At least one operator is organizing. Creative Skies published an industry alert on Docket 26-189 arguing that six of the seven categories describe aircraft show pilots never touch, and that Category 7’s second bullet reaches the entire light show business.

The FCC asks directly whether its list adequately captures military-grade equipment and whether domestic alternatives exist, and the alert lands on that opening: “They are asking us the exact question we can answer.” The warning in the alert is about the record. If the docket holds a national security determination and filings from defense-funded commenters, and nothing from the people who build and fly show aircraft, the proposal gets adopted on that record. Comments go through the FCC’s Electronic Comment Filing System under PS Docket 26-189, and the Drone Advocacy Alliance action page walks through both filing routes. Disclosure: DroneXL has been a Drone Advocacy Alliance partner since June 2024, and the Alliance receives sponsorship from DJI.

DroneXL’s Take

A show drone carries a light bulb. It flies a pre-programmed path inside a geofence, over a secured perimeter, under an FAA waiver that took months to win. Brightflight’s new 500-foot authorization runs five pages of provisions, down to NOTAM filing windows and a prohibition on ADS-B transmissions. This is the most supervised category of civilian drone flight in America, and the FCC just filed its aircraft under military capability.

Nova Sky Stories drones sit powered up on their launch pads in Central Park, New York, ahead of a 2023 drone light show. Photo credit: Haye Kesteloo / DroneXL

I will give the Commission the honest version of its case. Swarm technology is genuinely dual-use, and Damoda’s one-click container launcher unsettled defense analysts for defensible reasons. Chinese firms holding roughly 80 percent of the global show equipment market is a real concentration, not a talking point. If the concern is that hardware, the December action already answered it: no new foreign-built show drone can earn an FCC authorization without landing an exemption. The July proposal reaches past that border and into the replacement pipeline of American small businesses.

And look at who it actually catches. The workhorse of the US industry is built in South Korea, a treaty ally, by a company headquartered in Seattle that just raised American venture money. Write “produced in China” and you get a supply chain policy. Write “foreign” and you sweep a Texas operator’s Korean fleet into the same bucket as hardware from Shenzhen. That is not security drafting, and my read is the category was written to be wide first and justified second.

The question that decides most of the industry’s supply line is whether the Korean-built IFO can land one of the three exemption lanes: Blue UAS, Buy American, or a Conditional Approval with an onshoring plan. Nothing in the docket answers it yet. Watch the September 2 record for whether show operators put their fleet counts and replacement costs in front of the agency, because the FCC decides on the record it gets, and right now the loudest voices in that record want this hardware gone. An industry that replaced fireworks by being safer should not go quiet while being defined as a weapons category.

Sources: FCC Public Notice DA 26-758, Federal Register 91 FR 48870, FCC Fact Sheet, Creative Skies Industry Alert, DroneVideos, Yahoo Finance, The War Zone, FAA Certificate of Waiver 107W-2026-01798

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Saildrone Fires First Missiles From Its Robot Ship at US Navy's RIMPAC

Haye Kesteloo — Sat, 22 Aug 2026 20:24:00 GMT

A Saildrone Surveyor fired two Lockheed Martin Joint Air-to-Ground Missiles (JAGM) off the coast of Hawaii during the US Navy’s Rim of the Pacific (RIMPAC) 2026 exercise, the first missile launch ever conducted from one of the company’s wind-powered uncrewed vessels. Saildrone announced the demonstration on August 20, about three weeks after RIMPAC wrapped up on July 31.

Until recently, these hulls were better known for sailing into hurricanes for NOAA than for firing weapons. That résumé now includes live ordnance.

Two Missiles Left the Rail Through the Navy’s ARES Combat System

The Saildrone Surveyor fired both rounds while acting on target data for a surrogate high-speed surface craft, with the Navy’s Adjunct Remote Engagement System (ARES) handling weapons control. The launches took place in coordination with the USS Theodore Roosevelt (CVN-71) Carrier Strike Group.

ARES matters here. It is a remotely operated combat system, which means a human, not the robot ship, made the shoot decision. The 66-foot (20-meter) Surveyor supplied the platform, the persistence, and the sensor picture; the kill chain still ran through Navy hands. The demonstration was sponsored by the US Navy Fleet Experimentation Program, the Navy’s fast lane for testing new technology inside real exercises, according to Saildrone’s announcement.

Lockheed Martin bills JAGM as the in-production successor to the Longbow Hellfire, so the munition itself is thoroughly proven. What changed is what it was bolted to (a mission set Ukraine’s improvised sea drones forced onto every navy’s whiteboard). Richard Jenkins, Saildrone founder and CEO, said the test paired a trusted munition with “the most operationally deployed class of USV on the planet.”

Lockheed Martin’s October Investment Produced a Live Launch in Six Months

Lockheed Martin put $50 million into Saildrone in October 2025. Six months of joint engineering later, the integrated launcher fired live rounds at sea. The demonstration ran under the US Navy Fleet Experimentation Program, which sponsors rapid tests of new technology during fleet exercises.

The October partnership made Lockheed Martin the lead mission integrator while Saildrone kept shipbuilding. The RIMPAC event was the payoff both companies promised at signing: an on-water, live-fire demonstration in 2026, delivered with months to spare. Lockheed has been buying its way across the unmanned maritime domain at speed; DroneXL covered its Lamprey autonomous submersible in February, and the company closed a $3.45 billion deal for sonar maker Ultra Maritime in July.

The Surveyor did more than shoot. During the same exercise it deployed a passive electronic warfare sensor that identified a surrogate threat radar while working with an MH-60 helicopter, the kind of quiet listening job these hurricane-hardened hulls were doing for scientists long before anyone armed one. Paul Lemmo, Vice President and General Manager of Sensors, Effectors and Mission Systems at Lockheed Martin, said the companies will keep working to “invest in new capabilities ahead of need.”

The 170-Foot Spectre Carries the Bigger Payload Ambition

Saildrone’s 170-foot (52-meter) Spectre, the first of which is under construction, is designed to carry Lockheed Martin’s containerized Mk 70 launcher and towed sonar arrays for anti-submarine warfare, with a concealed deck sized for up to five 20-foot containers. RIMPAC was the warm-up.

Saildrone says the Spectre will displace around 330 tons (300 tonnes) fully loaded and reach 30 knots, with two variants: one with the company’s signature wing sail for near-silent transits, one without for a lower profile at speed. The anti-submarine ambition is not new territory for the company either. Saildrone was already testing towed sonar with Thales in April 2025, well before Lockheed’s money arrived.

A Saildrone uncrewed surface vessel, hull 3001, on a long-endurance autonomous mission in open ocean. Photo credit: Saildrone

The Surveyor that fired at RIMPAC stays in the fight as the smaller sibling: a multi-mission vessel built to hold station for months at a time, now with a JAGM dual launcher on the bow.

DroneXL’s Take

Extending the fleet’s missile magazine is the mission commanders most want from uncrewed surface vessels. Saildrone and Lockheed Martin just showed the shortest route there: bolt a combat-proven munition onto a hull family Saildrone says has logged more than 2 million nautical miles.

I don’t think the missile is the story here. The calendar is. A defense prime and a commercial drone builder went from signed deal to live rounds at a fleet exercise in well under a year, while conventional shipbuilding programs measure schedules in decades. Lockheed’s own write-up ties the effort to the Navy’s push for affordable mass by 2027, and this is what affordable mass looks like when someone actually builds it instead of briefing it.

Yes, this was a scripted shot at a surrogate target with a human on the ARES trigger and nobody jamming the link; combat conditions will be meaner. Saildrone has at least been operating in GPS-denied and spoofed waters in the Middle East, which answers part of that objection, though not all of it. The concept still holds. Ukraine proved cheap sea drones change naval math; the Pentagon is now buying that lesson in American form.

Watch whether a Mk 70 round leaves a Spectre deck before RIMPAC 2028. The first hull is under construction and the integration is already on both companies’ roadmap. If that shot happens, a wind-powered robot ship will be carrying the same missile cells as an Arleigh Burke destroyer.

Source: Saildrone

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DJI Mini 5 Pro Amazon Bundle Packs Over $100 In Free Gear For $899

Haye Kesteloo — Sat, 22 Aug 2026 19:28:00 GMT

An Amazon listing is selling the DJI Mini 5 Pro with a hardshell backpack, a high-speed 128GB microSD card, a 31.5-inch (80 cm) landing pad, and a 26-month extended protection plan for $899. Bought separately, those extras would run north of $100.

The bundle comes from a third-party seller pairing DJI hardware with its own accessory kit, and it is one of many. Search Amazon for this drone today and a rotating stack of similar listings appears, each adding memory cards, bags, landing pads, or warranty coverage on top of the aircraft. DroneXL has logged every price move on this drone since its Amazon debut in March, and the pattern holds: straight cash discounts on the drone itself are rare, so sellers compete on what they pack around it.

Get the DJI Mini 5 Pro bundle for $899 at Amazon

DJI Mini 5 Pro. Photo credit: Air Photography

The 1-Inch Sensor Is Still The Reason To Buy

The Mini 5 Pro is the first mini drone with a 1-inch CMOS sensor. It shoots 50-megapixel stills and 4K HDR video at 60fps, with 4K/120fps slow motion, 10-bit D-Log M, and up to 14 stops of dynamic range in auto mode.

The gimbal rotates 225 degrees and flips to true vertical, so portrait video for Reels and Shorts uses the full sensor instead of a crop. A 48mm medium-telephoto mode compresses backgrounds that a 24mm wide angle flattens. Front-facing LiDAR paired with omnidirectional vision sensors keeps obstacle detection working in low light, O4+ transmission is rated to 20 km (12.4 miles), and 42GB of internal storage covers the day you forget a card.

DJI rates the standard battery at 36 minutes. Treat that number as marketing: DroneXL’s long-term testing put real-world flight time at 20 to 25 minutes per battery, and nobody in our reviewer pool has matched the box figure in normal flying.

Air Photography And Jake Sloan Reached The Same Verdict

Two reviewers DroneXL publishes tested the Mini 5 Pro independently and landed in the same place. Shawn from Air Photography flew it for a month around launch. Jake Sloan ran it against the Mini 4 Pro in Alaska’s wind, cold, and ice caves.

Shawn reviewed the drone at launch and singled out its flexibility and video quality. “After flying the Mini 5 Pro for over a month, I’ve been truly impressed,” the Air Photography host said, calling it pro-level capability in a compact body that stays useful in places with tighter flight restrictions.

Jake Sloan tested both generations side by side in Alaska and found the image-quality gap smaller than the spec sheet suggests, with the two sensors trading wins on dynamic range. What sold him were the flight upgrades: better wind handling on mountain tops, quick-release propellers that skip the screwdriver in the snow, and the 2x crop for a telephoto feel. His Mini 5 Pro held a stable line through an ice cave with no GPS while his Mini 4 Pro lost positioning in the same cave and crashed. His conclusion: “the Mini 5 Pro is a huge upgrade and definitely worth it.”

Registration Applies No Matter Which Bundle You Buy

Every Mini 5 Pro sold in the United States needs FAA registration, whatever the listing says about 249.9 grams. Production units weigh 252 to 253 grams on a calibrated scale, and the FAA’s 250-gram threshold leaves no room for that variance.

A DroneXL industry source revealed that two last-minute design changes pushed production units past the printed weight. Europe’s EASA keeps the drone in its C0 class with a small buffer; American pilots get no such flexibility. Registration costs $5 at FAADroneZone, takes minutes, and brings Remote ID compliance along with it.

DroneXL’s Take

DroneXL logged the standalone Mini 5 Pro at $683 in July, its record low, after the base package held $759 for eight months. Fly More Combo listings have cycled through sales around $1,089. This $899 bundle sits between those markers, and its value depends on what is actually in the box, so check the configuration and the seller before checkout. Many of these bundles ship from third-party sellers rather than Amazon directly. We log every price move, with record lows dated, on our Mini 5 Pro deal tracker.

My read on the bundle wave is simple. With no official DJI sales channel in the United States and the FCC Covered List squeezing imports, sellers can’t compete on the drone’s price, so they compete on the accessory stack. The backpack, card, landing pad, and warranty are real, and they are also generic. If you were going to buy a bag, a card, and a landing pad anyway, a bundle like this saves you money and a few checkouts. If you weren’t, the standalone has already reset its floor once this year, from $759 in March to $683 in July. Deals on this drone recur. Panic buys don’t age well.

And whichever listing you pick, spend the extra $5 and register the aircraft. The scale says 252 grams no matter what the box says, and the pilot who skips registration is handing regulators their next talking point for free.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Amazon's Pool Drop Shows Why Zipline Is Winning Drone Delivery

Haye Kesteloo — Sat, 22 Aug 2026 17:04:00 GMT

An Amazon Prime Air drone dropped Lindsey Austen’s package into her swimming pool in Richmond, Texas, on Monday, August 17, and the video of the splashdown has been replayed on national news ever since. Austen ordered dry cat food and insoles. Twenty minutes later, an 83-pound (38-kilogram) MK30 hovered over her backyard, opened its cargo bay, and let the box fall into the water.

The timing stings for Amazon. Two days after the video was filmed, the company announced plans to bring Prime Air to nearly 500 US cities and towns by the end of 2026. I have covered Prime Air since the MK27 days, and the pool video captures the program’s core problem better than any spec sheet: the aircraft is enormous and loud, and it lets go of your cargo from 13 feet (4 meters) up.

Video credit: Lindsey Austen via Storyful

Amazon Built 83 Pounds of Aircraft Around a 5-Pound Box

Amazon’s MK30 weighs up to 83 pounds (38 kilograms) at takeoff and carries a maximum payload of 5 pounds (2.3 kilograms). The aircraft descends to about 13 feet (4 meters) and releases the box from a hover, which is how a package ends up in a pool.

Those figures come from Amazon’s own Prime Air operations page, which also lists a 7.5-mile (12-kilometer) range and cruise speeds of 60 to 74 mph (97 to 119 km/h). As I wrote after spending time with the aircraft at XPONENTIAL Europe in Düsseldorf this spring, that works out to 78 pounds of machinery hauling 5 pounds of cargo (the MK30 reaches my shoulder, and I am 6 feet 5 inches tall).

Amazon's booth slogan at XPONENTIAL Europe in Düsseldorf: "Gravity Shmavity." Photo credit: DroneXL

The record matches the mass. Two MK30s slammed into a construction crane in Tolleson, Arizona, in October 2025. In November, a Waco drone clipped a residential internet cable, drawing a second FAA probe. In February, one struck an apartment building in Richardson, Texas. And Austen’s package is the second Prime Air delivery to go swimming in a year, after an Arizona man watched his box get blown into his pool by the drone’s own downwash last August.

Video credit: CNBC Television

Amazon spokesperson Allie Payne told KHOU that incidents like Austen’s are uncommon and that “the response from customers using Prime Air has been overwhelmingly positive.” Austen sees it differently: “When it went into the pool, I was shocked,” she told Storyful. She says she is done with drone delivery.

Zipline Split the Delivery Into Two Aircraft

Zipline’s Platform 2 splits the job in two. The Zip hovers at roughly 300 feet (91 meters) while an autonomous droid rides a tether down, steering itself with small thrusters. It places up to 8 pounds (3.6 kilograms) in a drop zone about 2 feet (61 centimeters) across.

The Zip never comes near your patio furniture. The Platform 2 design keeps the big, heavy aircraft 300 feet up, where nobody hears it, while the duffel-bag-sized droid handles the last, riskiest stretch with a camera and its own sensors. Nothing free-falls. Jo Mardall, then Zipline’s head of engineering, told CNBC when the system was unveiled: “Success for us looks like being in the background, being barely audible.”

That philosophy is now operating at commercial scale. Zipline made its first Platform 2 customer delivery in January 2025, flies for Walmart across Dallas-Fort Worth, and signed a deal with Uber Eats this week that targets one million deliveries per day by 2029. Prime Air’s Detect-and-Avoid system is genuinely capable engineering. The difference is that Zipline engineered the delivery itself, the part the customer actually experiences, while Amazon engineered an aircraft and left the delivery to gravity.

Zipline’s Parachute Solved Rural Delivery First

A catapult launches Zipline’s original Platform 1, which cruises like a small airplane and drops its cargo under a parachute into an area the size of two parking spaces. A wire catches the returning aircraft, and Zipline puts its fleet at more than 125 million autonomous miles.

That fixed-wing design carried blood and vaccines across Rwanda and Ghana for years before American suburbs ever saw a delivery drone, and it still flies: Walmart uses P1 parachute drops in Greenville, Texas today. Two platforms, two problems, two purpose-built answers. Long-range rural delivery gets a glider and a parachute. Precise urban delivery gets a hovering mothership and a tethered droid. Amazon’s answer to both is the same 83-pound aircraft opening a hatch, and as our Düsseldorf analysis noted, Zipline’s and Wing’s delivery aircraft weigh between 10 and 40 pounds.

DroneXL’s Take

The pool video is being treated as a funny glitch. My read is harsher: it is the design working as designed. Every single Prime Air delivery ends with a 5-pound box in free fall from 13 feet, and whether it lands on a driveway, a hot tub, or a kid’s inflatable unicorn is a bet Amazon places hundreds of thousands of times a year.

To be fair to Amazon, that volume is real. Prime Air has delivered hundreds of thousands of packages in 2026, and a handful of viral failures against that denominator is a strong record by aviation standards. But scale cuts the other way too: multiply a drop-based delivery by the 500-town map Amazon just promised, and the pool videos become a genre.

Zipline never gives its customers that lottery ticket. The parachute floats. The droid sets the package down. Nearly 13 years after Jeff Bezos went on 60 Minutes, the company that started by flying blood bags across Rwanda has quietly out-engineered Amazon at the only moment that matters: the last 13 feet. Watch Amazon’s own end-of-2026 deadline for the 500-town expansion. If the pool videos keep coming as the map fills in, the MK30’s drop mechanism, and the size of the aircraft built around it, will be the reason.

Sources: Storyful via KHOU 11, Amazon, Zipline, CNBC

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

The FCC's 'Military-Grade' Drone Definition Is Broad Enough to Catch Your DJI Neo 2

Greg Reverdiau — Sat, 22 Aug 2026 15:13:00 GMT

On August 3, 2026, the FCC published a request for public comment that could reshape which drones are legally allowed to enter the US market. In plain English: drones the FCC has already approved could stop being sellable in this country. The agency is proposing to prohibit a category of equipment it’s calling “military-grade” — and as I’ll walk through here, that definition reaches a lot further than the name suggests.

This Is a Public Notice, Not a Rulemaking — and That Matters

First, an important distinction. This is not a Notice of Proposed Rulemaking. It’s a public notice issued jointly by two FCC bureaus — the Public Safety and Homeland Security Bureau and the Office of Engineering and Technology. The practical difference matters: the authority here has already been delegated to those bureaus, so they can act on the record they build without a vote from the full Commission. They aren’t even required to have a specific rule written at the time they publish the notice.

Some quick background if you missed it. On December 22, 2025, the Public Safety and Homeland Security Bureau added all foreign-produced UAS and UAS critical components to the FCC’s Covered List. Anything on that list can’t receive new FCC approval to be imported, advertised, or sold in the US. That addition rested on a national security determination from an executive branch interagency body.

This new notice goes further. It targets foreign-produced drones that were already authorized before that December 2025 cutoff — models a lot of you watching are flying right now — and proposes labeling certain ones “military-grade UAS,” which would prevent them from being sold going forward. To be clear: drones you already own and fly today aren’t affected by this notice. But if you break one, need parts, or want a replacement, this will affect you.

Seven Categories Define “Military-Grade”

The bureaus propose seven categories that would qualify a UAS as military-grade, each built on definitions borrowed from other agencies’ rules for high-risk UAS:

Heavy aircraft. Anything weighing 55 pounds (25 kilograms) or more at takeoff — outside the FAA’s definition of a small unmanned aircraft. The stated rationale is payload capacity, endurance, and range.
Agricultural sprayers. Aircraft capable of dispensing “economic poison,” the FAA’s legal term for pesticides and similar chemicals. This captures agricultural spray drones regardless of their weight.
Thermal imaging. Aircraft with sensors that capture and translate temperature differences, heat signatures, and residual heat signatures.
LiDAR. Aircraft with sensors that measure distance by illuminating a target with a laser and analyzing the reflected light.
Docking stations. Multi-purpose systems that let an aircraft land, take off, recharge, swap batteries, and transfer data and payload.
Defense-article integration. Aircraft specifically designed to incorporate a defense article. This one can probably be crossed off the list for most people watching this.
Swarming systems. This covers two things: ground control stations or flight-control systems purpose-built to manage drone swarms, and aircraft purpose-built to fly in coordinated, synchronized formation.

The label is “military-grade,” but the test is capacity-based, not use-based — a drone qualifies because of what it can do, not who’s flying it or what mission it’s actually on.

A Capability Test Catches a Lot of Civilian Equipment

That distinction has real consequences for three of these categories in particular. Thermal imaging is standard equipment for roof and moisture surveys, solar panel inspections, electrical utility work, and search and rescue. LiDAR is standard across surveying, corridor mapping, and forestry — and plenty of recreational drones use it for obstacle avoidance. And category seven reaches drone light shows: synchronized-formation platforms fit the swarming definition exactly.

What the FCC Is Asking the Public

The bureaus explicitly invite comment on seven questions. If you’re filing a comment, all seven are worth addressing:

Does the list get it right? Does it adequately capture military-grade systems — is anything on it not actually military-grade, and is anything military-grade missing?
What’s the real economic impact? The FCC currently expects this to be minor and contained, and wants data to back that up or refute it. Don’t just cite the cost of a replacement drone — the ecosystem cost matters too: new batteries, payloads, chargers, controllers, possibly new software, training, calibration, RTK equipment, docks, API integration, and downtime.
Is 180 days enough? The FCC proposes a 180-day implementation window after final action and asks if that’s sufficient. It’s also asking about inventory already imported and sitting on shelves, models in transit, and equipment under signed distribution or sales agreements — relevant if you’re a dealer or have a fleet on order.
Does the security benefit outweigh the economic cost? I’ll let you decide on that one.
Are there real domestic alternatives? Would switching actually be cost-effective? If you’ve researched an American-made replacement for your thermal or LiDAR platform, this is where it counts — but be specific about why it falls short: doesn’t support your sensor, less flight time, costs three times as much, six-month lead time. An alternative existing and an equivalent, commercially viable alternative existing are two different things. For recreational flyers and prosumers, there currently isn’t an alternative at all — say so.
Is the claimed economic upside real? The bureaus argue the ban would drive investment into domestic drone production, pointing to billions already raised and jobs already created since the Covered List addition. If your experience is that this investment hasn’t yet turned into products you can actually buy and fly, that’s worth putting on the record.
Do these systems pose an especially acute risk? Beyond the general Covered List finding, the FCC wants to know if military-grade drones specifically pose severe risk — unauthorized surveillance, data exfiltration, supply-chain vulnerabilities. Commercial operators can speak to how their data is actually handled: local storage, no cloud sync, firmware control, whatever applies. It’s a strange question to be asking after years of being told these risks were already proven — but here we are.

The Carve-Outs, and Where the FCC’s Own Logic Falls Apart

The proposed prohibition wouldn’t touch non-military drones, domestically produced equipment, drones on the Blue UAS Cleared List, or anything already exempt from the Covered List — which, once you apply the seven categories above, isn’t much. And to repeat: drones you already own stay legal to fly.

But here’s where the FCC’s reasoning starts to contradict itself. The agency expects minimal economic impact because, in its own words, military-grade drones are a minority of the market and recreational consumers are unlikely to be flying military-grade UAS. Except — remember how “military-grade” is defined here. Any aircraft that integrates a LiDAR sensor qualifies. And plenty of consumer drones use LiDAR for obstacle avoidance. Under the FCC’s own definition, the drone in your backpack right now is military-grade equipment.

That premise — that recreational flyers won’t feel this — is contradicted by the FCC’s own list, two pages earlier in the same document. You can’t write a definition broad enough to capture a sub-250-gram camera drone and then conclude everyday users won’t notice.

A DJI Air 3 sits behind a red no-entry symbol under a bold BANNED? headline, illustrating the ongoing debate over a potential U.S. ban on DJI drones. Photo credit: Pilot Institute

This proposal appears to reach every current DJI drone, from the consumer-grade Neo 2 up through the Agras agricultural line and the FlyCart delivery drones. It would also remove the ability to buy existing models from most of Autel’s lineup, and from manufacturers spanning Switzerland, Japan, and beyond — companies like Fotokite, Leica, and ACSL’s SOTEN — unless those companies eventually secure a waiver.

How to File a Comment

File a comment through the FCC’s Electronic Comment Filing System under PS Docket No. 26-189. Be specific, be personal, be professional — unprofessional comments and off-topic comments get ignored. Answer the questions above, and be specific about how this actually affects you. Comments are due by September 2, 2026, which is coming up fast.

You can watch the full episode on the Pilot Institute YouTube channel and find more of my coverage on my DroneXL author page.

FCC Tells All Certification Labs To Tighten Drone Screening

Haye Kesteloo — Sat, 22 Aug 2026 00:15:00 GMT

The FCC has told HOVERAir that importing, marketing or selling its VERSA flying camera in the United States is “a violation of U.S. law” and directed the company to stop immediately. The part of that statement with the longer reach was aimed at nobody in particular.

An FCC spokesperson told DroneXL the agency is “sending guidance to all Telecommunications Certification Bodies to strengthen their due diligence of applications” so they comply with their legal obligation not to certify equipment on the Covered List. That instruction goes to every body the Commission accredits.

The trigger was one bad grant. A certification body approved the VERSA, a Chinese-built camera that snaps into a propeller module and flies. The FCC says that authorization “was granted erroneously” and has since been set aside and pulled from the public database. By then the product had been announced as US-legal, covered as US-legal, and funded by American backers.

The FCC Outsources Nearly All Equipment Authorization

A Telecommunications Certification Body is a private firm the FCC accredits to review applications and issue grants of equipment authorization in its name. Manufacturers choose their TCB and pay it. The agency sets the rules, but the decision to approve a device sits with a contractor.

This is not an accident of the system. In an October 2025 order the Commission described TCBs as private third-party organizations to which it has delegated particular responsibilities, tasked with certifying that equipment complies with requirements both technical and non-technical, the latter category expressly including the rules prohibiting authorization of covered equipment. Screening the Covered List is their job, in writing.

The scale is the part worth absorbing. Compliance consultancy MarkReady counted 591 FCC-accredited test labs worldwide in April 2026, of which 67 also hold TCB status and can therefore both test a device and certify it. Fewer than a quarter of those labs sit in the United States.

The judgment they were handed is not an engineering one. Per the FCC, the VERSA camera on its own is “a prohibited UAS critical component produced in a foreign country” and, once the Flight Kit is attached, the whole thing “falls under the ban on UAS produced in a foreign country.” Reaching that conclusion means parsing a category, not measuring an emission.

The category is wider than it looks. The December 22 public notice defines UAS critical components as including, but not limited to, data transmission devices, communications systems, flight controllers, ground control stations, navigation systems, batteries, motors, and sensors and cameras. A camera sits on that list by name, which is how a handheld gimbal ends up prohibited with no propeller anywhere near it.

Where the category stops is a live question. The FCC’s own Covered List FAQ states that the nationality of the producing entity is not relevant, because the test is whether the item was produced in a foreign country. An open-ended component list combined with a country-of-manufacture test leaves an obvious boundary problem: at what point does a foreign-built camera that could be mounted to an aircraft become a UAS critical component rather than a camera.

For now that call starts with the manufacturer. The same FAQ places responsibility on applicants to certify in good faith that a device was not produced in a foreign country, with the certification body reviewing the claim. The screen the FCC now wants tightened begins as a company’s own assertion about its own product.

Bad Grants Have Reached The Market Before

This is not the first failure of the screen. On June 9, the FCC announced that its Office of Engineering and Technology had found 33 separate FCC IDs relying on identical test reports from a single lab, SLG-CPC. Confronted, the lab conceded it had submitted falsified reports.

Some of those reports covered entirely different products, and the lab blamed its own review procedures.

The agency has also been removing labs wholesale. Under the Bad Labs order adopted in May 2025, the FCC had withdrawn or denied recognition to 23 laboratories by April 2026, all tied to foreign adversary ownership or control. In April it voted to go further and restrict labs in countries without reciprocal recognition agreements.

Those actions targeted who owns the lab. The VERSA grant, where the FCC gave DroneXL its full statement, raises a different question, because nobody has suggested the certification body involved was compromised. It appears to have read a Covered List category and reached the wrong answer.

The FCC could pull the grant at all because it voted itself the power to revoke previously approved equipment in October 2025. That authority was built for hardware hiding Covered List components. It is now being used on a certification body’s reasoning.

The Drone Rules Have Been A Moving Target Since December

The FCC added all foreign-produced drones and UAS critical components to the Covered List on December 22, 2025. Since then it has carved out Blue UAS hardware, Buy American products, devices holding Conditional Approvals, and in June a category of toy drones capped at 150 grams.

It has also spent the year chasing evasion, proposing $25,000 forfeitures against eight alleged DJI front companies in July and later reaching additional DJI-linked brands. DroneXL has been documenting the gaps between the rule and its enforcement since January.

Every carve-out is another judgment call handed to a contractor with a paying client on the other side of the desk.

The Guidance Itself Has Not Been Published

The FCC has not released the guidance itself. There is no docket number attached to it, no public notice, and no indication whether it will appear as a formal document or an internal advisory. The agency has also not named the certification body that issued the VERSA grant.

The Commission’s own Covered List page logs every related document it has issued, running through a power inverter notice on August 20. Nothing about TCB due diligence appears there.

That gap matters for a practical reason. A manufacturer with a pending application cannot comply with guidance it has not seen, and a TCB weighing a borderline drone-adjacent device this week has the same Covered List text it had last week. DroneXL has asked the FCC whether the guidance will be made public.

DroneXL’s Take

I have spent eight months arguing the Covered List is bad policy for American operators. This story is about something narrower and, in its way, more troubling. The rule is not actually being enforced by the government that wrote it, and the VERSA grant is how we found out.

The FCC wrote a categorical ban on an entire class of hardware, then delegated the call to roughly 67 private certification bodies scattered across the globe, most of them outside the United States, all of them paid by the applicants whose products they judge. It discovered this bad grant not through an audit but after a company had banked crowdfunding money and the tech press had told readers the thing was legal. The remedy is a memo.

I want to be fair about the difficulty, and about one fact that cuts against me. The Commission did not quietly assume TCBs would handle Covered List screening. It wrote the duty into an order in October 2025, in plain language. A regulator with a few hundred engineers also cannot personally examine tens of thousands of devices a year, and delegation is how every modern conformity assessment system works.

But a duty on paper is not a screen in practice. If a category is written so that a competent, uncompromised lab can look at a camera with detachable propellers and approve it in good faith, the category is the problem, not the lab. The agency now says HOVERAir was “apparently willfully violating” the rules, which may be true, and which also raises the question of how a willful violation cleared a body whose stated job was to catch it. My read is that the VERSA is not an outlier but the first one caught, and that the same ambiguity is sitting inside other grants issued since December.

The test is whether the guidance ever becomes a public document with a docket number. Guidance you can read is a rule. Guidance you cannot is a press statement, and this industry has had plenty of those.

Sources: FCC spokesperson statement provided to DroneXL, FCC Public Notice DA 25-1086, FCC Covered List FAQ, FCC Order FCC 25-71, FCC Office of Engineering and Technology, MarkReady

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

The DJI Osmo Pocket 4 Accessories I Actually Use: Protection, ND Filters, Mics, Batteries, and Car Rigs

Jake Sloan — Fri, 21 Aug 2026 18:39:44 GMT

If you just picked up the DJI Osmo Pocket 4 and you want to protect the camera, take your footage to the next level, or just in general make it easier to use, I have a few accessories that I want to show you. Now, some of these accessories were sent to me, and some of them I purchased myself — I’ll try to let you know which ones are which. And if you’re interested in picking any of these up, there are affiliate links down in the description of my video. They don’t cost you anything extra, but they do help me out.

Protect the Gimbal and Screen First

Depending on which kit you get with the Pocket 4, you might get a little bag, which is actually really nice and does a good job of holding all your accessories, but it doesn’t necessarily protect the camera and the gimbal. That’s where a protective cover comes in. This is something I purchased off Amazon, and it does exactly what you need it to do, which is protect the Pocket’s gimbal, camera, and front screen if you’re just going to throw it in your camera bag. It clamps on, comes around the back, and snaps into place, keeping the screen and buttons safe and the camera nice and secure, so you can just toss it in your bag or backpack. It took me a while to find, but I’m definitely glad I did. It doesn’t add much size or bulk to the camera, but it gives you a nice way to protect it, which is important.

Sort of related to that, but definitely something you’re going to want to pick up, is a screen protector. I’ve got mine on and you can’t really tell it’s there. It’s just a little piece of glass, but it will save your screen if you drop this thing in just the right way. Super cheap, but very, very highly recommended.

A compact handheld gimbal camera sits in its protective housing, surrounded by the straps, mount and accessories it ships with. Photo credit: Jake Sloan

A Magnetic Arca-Swiss Base That Does Everything

You get a little tripod and adapter with the Pocket 4 no matter which package you buy, but if you want a little more versatility, there’s this K&F Concept base, which is super cool. Not only is it an Arca-Swiss plate — so if you use a regular tripod, you can clip it straight into your Arca-Swiss clamp — but it’s also magnetic, so you can stick it to anything metal. This is something I purchased, and pulling it off does take a little work because it’s new.

It also has a little tripod built in, so if you want to set your Pocket 4 somewhere and let it just sit, you can do that. You can also raise a little piece that lets you hook it on a fence or a bookshelf if you want to. This is probably one of the most versatile little Pocket 4 base adapters I’ve found. Really nice and, like I said, very highly recommended.

A DJI Osmo Pocket sits on a compact K&F Concept tripod during a workbench test setup. Photo credit: Jake Sloan

ND Filters for the Highest Quality Footage

Something I picked up and have been using a lot more than I thought I would: if you want the highest quality footage from your Pocket 4P — and because it has D-Log 2 and D-Log, why wouldn’t you want the highest quality footage you can get out of it? — you’re going to want ND filters. The camera is actually pretty dang incredible. These were sent to me by Freewell, and I use them a ton. The everyday pack is excellent. It has a lot of filters that you would need for pretty much anything, and if you’re going to shoot a lot of general video, that’s where I would start.

A full Freewell ND filter set sits in its magnetic folding case, with ND, polarizing and Glow Mist 1/4 elements ready for a drone camera. Photo credit: Jake Sloan

If you want to step up and do other things like time lapses, hyperlapses, or car rig work, then I would get the kit that includes a CPL, the light pollution reduction filter, the ND1000, and the Glow Mist filter. The ND1000 is super useful for time lapses or hyperlapses because it allows you to slow your shutter speed way down and blur lots of motion. The mist filter is super cool if you’re shooting around a lot of strong light sources, like in a city — it softens the light sources and makes them bloom a little bit for a really cool effect.

A full set of ND and CPL filters sits in its magnetic carry case, ready to be swapped onto a drone's camera between shots. Photo credit: Jake Sloan

Freewell also makes a combination polarizer and ND filter, which is useful more for photography. You can control the contrast in the sky and in the leaves, and you can keep reflections off windows — so if you’re shooting through a car window or something, polarizers are very important for that. If you’re doing more photography, or you’re really concerned about reflections because you shoot a lot around water, that ND and polarizer combination is worth considering.

A circular polarizer strips the reflections off the windshield, leaving the driver and passenger clearly visible inside the orange sports car. Photo credit: Jake Sloan

The Wide-Angle Lens for Walk-and-Talk

The main camera on the Pocket 4P is 20mm, and that’s great for a lot of things. But if you’re going to do a lot of shooting where you’re holding the camera, walking, and talking to it, you’re probably going to want the wide-angle lens. You might get it depending on the combination you buy, but if you don’t, I would highly recommend it. It makes the shot just a little bit wider, a little bit easier to walk and talk to the camera without feeling quite so tight, and it shows off a lot more around you.

A dual-lens pocket gimbal camera swivels on its stabilized head during a field test on a wooded trail. Photo credit: Jake Sloan

A hiker carries a yellow pack with a tripod mounted on the side along a trail through a mossy Pacific Northwest forest. Photo credit: Jake Sloan

A Microphone Makes People Stick Around

On that note, if you’re walking and talking to camera a lot, or you’re vlogging a lot, then you’re going to want a microphone. External mics give you a lot cleaner audio, they will really help you out when it’s windy, and you can connect up to two mics to this camera — so you could have two people talking directly to camera with really good, clean audio.

A DJI wireless mic transmitter with its deadcat windscreen attached, held above a creator's desk alongside a camera, carrying case and keyboard. Photo credit: Jake Sloan

The nice thing is that the Pocket 4 works with the DJI Mic Mini, the Mic Mini 2 and Mic Mini 2S, and the DJI Mic 3, plus the DJI Mic 2 if you want to do a lapel setup. So you have a lot of options there. Having an external microphone will really help you get much cleaner, much better audio, which in turn will help people stick around and actually watch whatever you’re creating.

Batteries: The Handle Works, but This Power Bank Is Better

Battery life on the Pocket 4P is actually really good, but if you want longer battery life or you’re shooting longer days, the Pocket 3 battery handle will work with the Pocket 4. You can pick up the Pocket 4 battery handle in a lot of places, and that’ll be a good option as well.

A DJI Osmo Pocket gimbal camera shown with an extended battery handle and a magnetic 3x tele lens fitted over its 1-inch sensor. Photo credit: Jake Sloan

But if you want a lot of battery life, this battery from SmallRig, which is something I purchased, is really great and gives you a ton more. It charges differently than the Pocket series batteries do — it works more like a USB power bank. The camera won’t tell you how much power is left in it, but it will just dump power into the Pocket 4 and recharge it. The way I’ve used it: when I’m going from location to location, I’ll plug the Pocket 4 into this battery, and by the time I get wherever I’m going — because the Pocket 4 and 4P have quick charge — it’ll be full, or at the very least up to 80 or 90%.

The LUTs I Grade All My Pocket Footage With

If you shoot in D-Log or D-Log 2, I use and highly recommend the Gamut.io LUTs for the DJI Pocket 4 and 4P. They are fantastic. That’s what I’ve color graded all of my Pocket 4 and 4P footage with in any of the reviews or comparisons I’ve done. They give you a really nice, natural-looking skin tone and just really nice colors. I can’t recommend them highly enough.

Yellow cinquefoil blooms catch the light on an exposed ridge, with a forested mountain slope falling away in the background. Photo credit: Jake Sloan

He sent me these early on when he was testing them out, and that’s how I was part of testing them for him. But then I loved them so much I went and bought them for my Mavic 4, my Air 3S, my Insta360 Luna — I bought them for just about every camera I have, because they are really fantastic-looking LUTs that give you great skin tones and great colors across the board. They also give you a couple of options between a slightly more toned-down look and a very bold look, so you can choose which one you want.

Car Rigs: The Pocket 4 Is a Fantastic Rolling Camera

I do have some other accessories that I haven’t tested yet — like an underwater case for the Pocket 4 and Pocket 4P that I bought, and I’ve got some ideas on how I’m going to try that out. But this one I have tested: the Movmax Pocket Controller, which works really well. If you do a lot of rolling stuff — car content, car rigs, anything like that — the Pocket 4P is actually a fantastic rig for it.

A MOVMAX crane arm and gimbal camera rig mounted to a camera car, set up for a low-angle tracking shot of a red Ferrari on a city street. Photo credit: Jake Sloan

I’ve used the Movmax Blade Arm, the little bouncy fourth-axis arm, and it works really well — that’s what all of this footage was shot with. For myself, I then went and bought the Tilta Hydra Alien Mini, just because I like a few things about it a little bit better. But if you’re going to do any car content, I would highly recommend getting one of these arms along with the Movmax Pocket Controller system, because it works super well for filming car rigs.

A red Ferrari California crosses a steel-grate bascule bridge at dusk, headlights on as the city skyline fades into the evening sky. Photo credit: Jake Sloan

The Pocket 4 — these little gimbal cameras, period — make a great car rig because they’re so fast and easy to set up, and they’re so unobtrusive that you can get away with filming in a lot of places where a huge car rig leaning out the side of a car would attract more attention than you want.

A Motorized Slider That Unlocks Something New

And I will have videos on this coming, but there’s one more: the iFootage Shark Slider Pico Pro. It gives you some really cool abilities to control not just left-and-right motion but also panning, and it interfaces with the Pocket to give you control over the gimbal. A whole video will be coming on this really soon, because it unlocks some really, really cool things you can do with your Pocket 4 or Pocket 4P camera. So stay tuned for that.

A compact motorized camera slider gets a close-up walkthrough in a home studio filled with gimbals, cameras and production gear. Photo credit: Jake Sloan

Let me know in the comments what accessories you use that I didn’t get to. As always, if you have questions, ask me in the comments, or you can join my live stream, which happens most Wednesday nights at 4:00 p.m. Alaska time, 8:00 p.m. Eastern. I’ve got to go out and film some more, because a video’s coming and it’s really cool.

This article was written based on Jake Sloan’s DJI Osmo Pocket 4 accessories video, published on his YouTube channel. You can find more of Jake’s coverage on his DroneXL author page.

Insta360 X6 In-Depth Review: What's Actually New — and Where It Falls Short

Ray DC Rainmaker — Fri, 21 Aug 2026 18:30:00 GMT

So, this is the brand new Insta360 X6. I’ve been putting this camera through its paces over the last month and a half across all sorts of different scenarios — underwater, on the water, on the ground, you name it — all to figure out where it works well and where it falls short. In this review, I’m going to dive into what is actually new. Not just the marketing junk they want you to hear, but whether or not these new features actually work and where some of them fall short, because not all of them are as good as they might seem.

As always, no company is paying for my reviews, nor do they have any rights to view them beforehand. That is super critical in this space. With that, I’ll simply dive into it.

A Shorter, Wider Body — and On-Screen Buttons I Don’t Love

The very first thing is that this camera is a little bit shorter than its previous generation. It’s also a smidge wider than the past. It’s not quite as short as the GoPro or DJI cameras, but it’s also not as wide. I like the new size, but I could also use it just a little bit smaller if we can get there.

The Insta360 X6 sits noticeably shorter and wider than the X5, with a repositioned lens and a live preview showing 8K/50 capture next to the X5's 8K/30. Photo credit: DC Rainmaker

Now, with that smaller case size comes a smaller screen than in the past — you can see it side by side with the X5 as well as the other cameras. I’m mostly fine with the new screen itself. But you may have noticed that they removed the buttons below it compared to the X5. Instead, these buttons basically show up along the bottom of the screen, and then you use a force touch system. There’s now haptic feedback — or basically a vibration motor, as it’s more commonly known — so you can feel when you’re pushing down on the screen. This isn’t great in my opinion. I think it’s something that’s fine, but it’s not as good as just a real physical button. You still have buttons on the side, but again, more buttons is better.

The last of the external things I’m going to talk about before we dive into the insides is the bottom. There’s a regular tripod mount that’s always been there, but you’ll notice there are now contact points. Those contact points are power. So you can basically have a power mount — there’s a powered version of the selfie stick as well — that feeds battery into the Insta360 X6. Super cool addition.

A close look at the underside of a rugged action camera, where the threaded tripod socket and magnetic mounting contacts sit flush in the textured housing. Photo credit: DC Rainmaker

New Replaceable Lenses — and the One I Scratched Anyway

The final external thing is that the company says they’ve improved the hardening on these lenses. These are replaceable lenses, just like they had with the X5. They are not compatible with the X5, though — different lenses. The new lenses are supposed to resist scratches better and are also cheaper.

The Insta360 X6 (right) sits beside the X5 (left), revealing a larger lens assembly, a textured body and a taller touchscreen on the newer 360-degree camera. Photo credit: DC Rainmaker

That said, I did manage to scratch one of my lenses on a very simple drop from a little table down to the ground. It’s like a foot and a half or so, and it just simply slipped off the table, hit the ground, and scratched the lens. The surface is a faux wood — it’s actually tile that looks like wood. I’ve also dropped this camera many other times, and it hasn’t scratched. So take that for what you will.

Bigger Sensors: How the X6 Stacks Up Against GoPro and DJI

The biggest internal improvement, if you will, is the change in sensors. They’ve gone from a 1/1.28-inch sensor to a 1/1.1-inch sensor — essentially the same sensor as in the DJI Osmo 360 camera. And again, these are dual sensors, because there are of course two different lenses.

Side-by-side sensor size comparison of the leading 360 cameras, with the Insta360 X6 and DJI Osmo 360 both at 1/1.1 inches, the Insta360 X5 at 1/1.28 inches, and the GoPro Max 2 trailing at 1/2.3 inches. Photo credit: DC Rainmaker

So let’s look at how this compares side by side to a couple of other cameras. Starting off in bright conditions, the X6 is definitely overexposing a little. The GoPro and DJI cameras look fine to me here — they’re pretty much spot on where I would expect.

Three 360-degree action cameras mounted on a single handheld rig for a side-by-side field test in the woods. Photo credit: DC Rainmaker

In another shot on the bike, the overexposure at first is so subtle you don’t even notice it. It’s only once I start panning around the front and my subject gets back into the brightness that you see the overexposure in the Insta360 X6 shot.

A three-way split-screen test pits the DJI Osmo 360, Insta360 X6 and GoPro Max 2 against each other on the same sunlit mountain road ride. Photo credit: DC Rainmaker

Meanwhile, earlier in the morning on the jet skis, where it’s not so bright, all of these look pretty good across the board — very, very similar.

A three-way image quality comparison pits the DJI Osmo 360, Insta360 X6 and GoPro Max 2 against each other on the same jet ski run along a Mediterranean coastline. Photo credit: DC Rainmaker

In a bunker-type location with graffiti in the background, we see a bit more depth and contrast in the GoPro footage than the X6 footage, and the DJI kind of splits the difference between the two. Another shot is interesting: in that case, the GoPro coloring is simply not correct. I would say the X6 coloring is maybe a little bit darker and more contrasty than it really is in real life, with the DJI managing to stick the landing in the middle.

The thing to remember in single-lens shots is that both DJI and Insta360 allow 5K, but GoPro only allows 4K in single-lens mode — despite the fact that GoPro actually captures more pixels in 360 mode. If you zoom in in post-production, you can clearly see the difference between DJI and Insta360 in single-lens mode. But the kicker is that if I switch over to 360 footage instead, where the GoPro is actually shooting more pixels — a true 8K frame versus the other two at 7.2K — it’s more or less a wash between these on tree bark zoomed in to 400%. Also notable: the color is more correct on the GoPro and Insta360 than the DJI on that one, for whatever reason. These shots were shot back to back.

10-Bit, I-Log, and Dolby Vision

The next item on the list: there’s finally 10-bit recording in their 360 cameras. We saw Insta360 add it to their Ultra Pocket series this past spring and summer, and now we’re seeing it come to other cameras, including this one. You simply swipe down the menus and enable 10-bit there. That includes I-Log footage, by the way, up to 8K50, though there are some minor limitations around that.

Next, they’ve added Dolby Vision recording to the Pro settings — basically something you can swipe in and toggle. I would say this is mostly a shrug in real life. This is a great spec thing, right? It’s a thing they’ve put all over the marketing. In real life though, nobody actually supports it for the most part. There is a handful of people who will find this useful, but most people will be like, shrug.

The New 4nm Chipset: 8K50, 5K Single-Lens, and InstaFrame 2.0

Instead, one thing that’s seemingly minor but is probably a bigger deal is that they’ve changed from a 5-nanometer chipset down to a 4-nanometer chipset. And with that, they’re lighting up a bunch of things that happen on the device itself.

First off, they’ve gone from 8K30 up to 8K50 for the frame rate on the 360 recording side. You’re also seeing the single-lens footage increase dramatically from 4K up to 5K. This means you can record a single lens at a time for a flat view. For example, if you have it on a chest mount, there’s no reason to record 360 because you can’t see my chest — well, you can anyway. Might as well just save the battery and record one lens at 5K60. Or you can do 4K120 frames per second if you want to capture something in slow-mo.

A 360-degree action camera's touchscreen settings menu, with 5K resolution and 30 frames per second selected and horizon lock listed as not supported. Photo credit: DC Rainmaker

One of the things they’ve done with this increased power is upgrade to InstaFrame 2.0. That’s a marketing term that means you can record both a 360 shot as well as a flat file shot, which simplifies the editing workflow quite a bit. In the past, that flat file was limited to 1080p, which isn’t super awesome, and the 360 was also limited to roughly 5K. Now, the 360 backup footage, as I call it, is all the way up to 8K30, and the main flat file footage is at 4K — basically four times the resolution compared to 1080p, which is a huge, huge leap.

This is probably what most people should be recording in, because it basically just makes your life a whole crap-ton easier. You just go out, record what you want, and in theory it gets the right shot. And if it doesn’t get the right shot, then you go back to the 360 footage you would have had to sift through anyway.

A reframed still from unedited Insta360 X6 8K/50 footage shows the camera's 360-degree capture holding a runner steady in frame while the full clifftop cove behind him stays sharp. Photo credit: DC Rainmaker

AI Director Now Runs on the Camera Itself

In fact, talking about sifting through footage, one of the things Insta360 has added to this camera is their AI Director. This is something you swipe down and toggle on the camera itself, and all the processing happens on the camera. You know how in the past Insta360 would upload all your footage to the cloud and it would spit out a little video summarizing your day? Now, that happens on camera. It can still happen in the cloud if you want, just like other cameras from GoPro and so on. But otherwise, it happens on camera, and it’ll upload a summary of your day — a little vlog, if you will. You can choose whether you want to prioritize battery charging or prioritize getting it done faster.

Low Light: Brighter, But Not Always Correct

Going back to the sensors, we’ve got to talk about low-light footage. Insta360 has claimed significantly better low-light footage with this camera over the X5, due in part to the sensor upgrades as well as some of the processing it can do.

The best way to see the difference between the X5 and the X6 is switching back and forth between the X5 footage and the X6 footage. Obviously, it is brighter on the X6, but it’s actually not as correct from a color standpoint. There’s a blue bus in my footage that is genuinely blue — it’s literally called the Blue Moon bus. It is blue, not green like the X6 thinks. Just like the bus, the detail is definitely far better on a campfire at night, and the coloring in that shot is better as well.

A low-light night test pits the Insta360 X6 against the X5 and an iPhone 17 Pro on the same lamp-lit deck. Photo credit: DC Rainmaker

If I go the complete opposite direction into Times Square, both DJI and Insta360 expose very brightly, because it’s a very bright scenario. The GoPro is a bit dimmer, because the Max 2 isn’t as good in low-light mode.

Three 360-degree cameras record side by side in Times Square, putting their low-light performance to the test against the neon glare of Manhattan billboards. Photo credit: DC Rainmaker

That becomes more evident as I walk down a much quieter street, where the Max 2 starts to struggle in the lower light. The DJI camera does seem to be doing some funky polish stuff on the ground that’s not real in real life, whereas the X6 seems to stick the landing best there. On the other hand, looking at my face, I would say the DJI actually does a better job of getting my face colors correct, while the X6 is again overexposing a little bit.

Low-light selfie footage from the DJI Osmo 360, Insta360 X6 and GoPro Max 2 compared side by side on a city street at night. Photo credit: DC Rainmaker

Underwater With the New Dive Case: A Bit of a Situation

Since we’re on the footage evaluation train, let’s go underwater — it’s probably prettier there anyway. The new Insta360 case for the X6 is obviously different because it’s a different-sized unit, and when it comes to underwater cases, they have to be super perfect for a given unit.

An action camera locked into a clear waterproof dive housing with a domed lens port, held over a swimming pool ahead of an underwater test. Photo credit: DC Rainmaker

But it’s more than just the case itself. The new underwater case is a situation, if you will. When you use it, you also have to swap out your lenses for underwater lenses that are designed for that particular case. That means you can’t really do it on the boat or wherever you’re going underwater, because from a humidity standpoint, you want to do that in a really low-humidity environment — typically indoors — so you don’t get any fogging. Even by putting the little anti-fog inserts in there, you’re still going to want to do this somewhere indoors before you head out on the water for the day. Which means your camera is really going to be dedicated to that case for the entire day.

Setting that aside, once we go underwater, the shots do look nice. Though, from a stitching standpoint, you can clearly see the stitch line across one of my shots with my wife. If I hold the camera up facing one lens, there’s no stitch line, so it’s nice and clean — but then you lose that 360 effect where the whole pole disappears on you. In talking to Insta360, I sent them a little tidbit of footage, and they said it’s just normal for that particular lens in that particular situation.

A freediver films underwater with an Insta360 X6 on an extended selfie stick, capturing 8K/50 360-degree footage over a shallow seagrass meadow. Photo credit: DC Rainmaker

It’s also a good time to mention that the underwater resistance has been improved from 15 meters to 20 meters, or 66 feet. But that doesn’t really matter too much, because at the end of the day, if you do not have an underwater case for this, it’s all going to be fuzzy underwater. That’s just the way the optics work for these lenses.

Internal Storage and a Completely Redesigned Battery

Next on the list, a different direction entirely: there is now internal storage in here — 47GB of internal storage compared to zero in the past. This follows what we’ve seen DJI do in their newer cameras. I appreciate that you can still put a microSD card in the side, as I do. But having that internal storage is just handy as a backup in case you popped out the SD card, forgot to put it back in, and went out on your adventure — it’s always there. I’m a huge fan of having a little bit of internal storage inside.

A 360 camera's storage management screen shows internal memory nearly full at 45GB of 47GB while the microSD card still has plenty of room at 108GB of 476GB. Photo credit: DC Rainmaker

Next up, they’ve redesigned the batteries entirely. Now you have a battery door. In the past, the whole door came off with the battery; now the battery is simply inside there. And it is a beefcake of a battery — a bigger battery than in the past, now 2,600mAh compared to 2,400mAh. This is such a much better design than having the battery as part of the door itself, so I’m really glad they switched in this direction. These batteries are not the same as their Ace Pro 2 cameras, though. Perhaps the Ace Pro 3 will have the same batteries — I hope so, Insta360.

The Insta360 X6 uses a new 2600mAh, 10.17Wh rechargeable battery that slots into the camera's side door. Photo credit: DC Rainmaker

In the meantime, the battery claims have increased quite a bit. They’ve gone from 93 minutes in 8K mode up to 140 minutes in 8K mode. That’s 8K30, no 10-bit recording. My testing results support that claim as well. Insta360 notes that in cooler temperatures you should get better battery life — not cold, because that’s a different realm of battery configuration stuff, but in terms of fall temperatures, it’ll probably do better.

120-Megapixel Photos and AI Reframing

Now we’ll round home with a bunch of photo stuff and a bunch of non-photo random things that are still cool. On the photo side, they’ve increased the megapixels from 72 megapixels up to 120 megapixels. I’ll drop some photos in a share down in my video description, as well as some footage, if you just want to fuss around with it.

Next on the photo side, they’ve added a new AI reframing automatic thingamajig, and it’s actually pretty interesting. The idea is you can take a 360 shot — I just hold the camera up, it takes a shot, boom — and it’s going to give you three different photos plus the original 360 photo. Of those three photos, one is going to be facing you, the subject, automatically reframed toward you. One is going to face the other direction, basically. And one is going to give you that little tiny-world view that you might use once a year before your friends get annoyed with you.

This is a cool concept, and it kind of simplifies some of that workflow. In the past, you would have had to take the 360 photo, reframe it, and so on. This is all about streamlining things. And in fact, I would argue it’s that sort of tiny feature that embodies probably the biggest difference between Insta360 and GoPro and DJI. Those two other companies both have cameras that, from a hardware standpoint, are basically in the same ballpark. GoPro doesn’t have as good low-light footage, but for the rest of the hardware specs, it has proper 8K footage versus this really being 7.2K footage when you do the math. That hasn’t changed here. Instead, Insta360 leans heavily on the software side of things. There’s a gazillion little features like that which Insta360 has and GoPro and DJI just don’t. And those are the sort of things that, for most users, make their life a lot easier.

NFC Tap-to-Connect, Faster Bluetooth, Wi-Fi 6, and USB 3.0

I’ll give you two more features in the general category of usefulness — they’re actually somewhat hardware, not software. The first one is tap-to-connect with NFC. You can basically just tap your phone on the back of the camera to connect to it and start downloading.

Next, they’ve worked heavily on the Bluetooth side of things, so that connection time is one second. I would say that’s optimistic in real-life testing here — it just tends to take longer, with the phone thinking about life, than one second. I’m guessing at the internals level it’s one second; at the software-and-everything-else level, it’s like five to seven seconds.

In the same speeds department, they’ve gone up to Wi-Fi 6.0, which gives you up to 100 megabytes per second of transfer speeds, and USB 3.0, which gives you 450 megabytes per second. That Wi-Fi is how you offload footage from the camera to your phone or up to Insta360’s cloud, and you can use USB if you want to do the same thing at faster speeds to a desktop. And of course, you can still pull the microSD card out of the bottom.

Pricing — and the DJI Osmo 360 II Waiting in the Wings

When it comes to pricing, the X6 starts at $699.99 for the Standard Bundle and $799.99 for one of the Essentials Bundles, and there are lots of other bundles from there. Replaceable lenses are $26 for a single and $32 for the dual. Obviously, you just get the dual, and then you get the batteries from there.

A pricing slide from an in-depth Insta360 X6 review lists the $699 standard bundle, the $799 essential bundle, and accessory costs for replacement lenses and batteries. Photo credit: DC Rainmaker

I will have a full comparison video coming between this and the GoPro and DJI models. Obviously, DJI is planning to crash this party with the DJI Osmo 360 II. But if we know anything about DJI’s launches in the last year or so, those of us outside of China won’t see that for months at best — and if you’re in the US, you won’t see it at all unless you go through some random place that will ship it to you. And it seems like the vast majority of the changes there are basically about finally having replaceable lenses like their competitors. But that’s a separate video for a separate day.

Should You Upgrade From the X5?

So, what do I think overall? Well, I think it depends on how you use your Insta360 camera or other 360 camera. If you have, for example, the X5 that shoots at 30 frames per second, the bump to 50 frames per second probably isn’t that big a deal. And while the low-light footage is definitely improved, it’s also not a huge jump like you might have seen from some other cameras.

On the flip side, for things like InstaFrame — the option where you can shoot a flat file version while having the 360 backup — that is a massive upgrade. Going from 1080p to 4K for the flat file, and now to 8K for the 360 footage, is a huge jump for that particular feature. Same goes if you plan to use 10-bit recording — that’s a big jump you wouldn’t have had in the past.

Long story short: if you’re pushing the boundaries of the X5 today, wanting things like 10-bit or wanting to have the backup in 8K while also doing a flat file, then definitely jump to the X6. But if you’re happy with the X5 today, largely shooting at 8K30 and just snipping out the parts you want manually, I would probably stay with that.

Either way, it is cool to see Insta360 doing notable things on the hardware and the software front year after year. They could have easily just sat back and done nothing — maybe changed the replaceable lenses and called it a new version. But they didn’t. They added a bunch of new things here to keep pushing that envelope forward. I’ve got a full comparison video coming out comparing a bunch of these 360 cameras at once on tons of different footage, so stay tuned for that.

This article was written based on Ray’s Insta360 X6 review, published on the DC Rainmaker YouTube channel. You can find more of Ray’s coverage on his DroneXL author page.

DJI Offers $10,000 For Your Best Avata 360 Video While Washington Decides Its US Future

Haye Kesteloo — Fri, 21 Aug 2026 17:56:00 GMT

DJI is paying for footage again. The drone maker’s All For Passion Challenge offers a $10,000 grand prize for the best sports or action video shot on the Avata 360, its 8K 360-degree FPV drone, with entries open through September 30 and winners announced October 28.

The full pool is $25,000: one $10,000 grand prize, two $5,000 runner-up prizes, and five $1,000 third-place awards. DroneXL has tracked this drone since the first promotional shots leaked last November, and a creator contest was the predictable next move for a product DJI built around reframable, share-ready footage. What makes this one worth a closer look is less the money than the machine underneath it, and what it took to get that machine into American hands.

DJI Puts $25,000 In Prizes Behind The All For Passion Challenge

The All For Passion Challenge awards one $10,000 grand prize, two $5,000 runner-up prizes, and five $1,000 third-place awards for sports and action videos shot on the DJI Avata 360. Entries close September 30, and DJI will announce the winners on October 28.

The mechanics are simple, according to DJI’s contest announcement. Film a sport with the Avata 360, add the official “DJI AVATA 360 | ALL FOR PASSION” watermark, and publish on Instagram, YouTube, Facebook, or TikTok with the hashtags #allforpassion, #djiavata, and #avata360. Then email the post link, platform, and publication date to global-creator@dji.com. There is no cap on entries, so a productive summer can produce a stack of tickets.

Eligible activities run from mountain biking and surfing to skiing, climbing, football, and swimming. If it moves fast enough to be worth chasing with a cinewhoop, it qualifies.

DJI has been seeding the campaign since at least July 23, when it announced Sports Series Stickers for the challenge on X. And for scale: this is pocket change next to the SkyPixel contest and its prize pot of almost $200,000, but SkyPixel spans DJI’s whole catalog. This one exists to sell exactly one aircraft.

Entries Compete On Quality, Not Follower Count

DJI will judge entries on three criteria: video quality, creativity, and influence. Follower counts are not one of them. Participants keep copyright, but entering grants DJI a worldwide, perpetual license to use, edit, and share submissions for promotional purposes with attribution.

That license language is standard for manufacturer contests, and it is still worth reading before you upload your best work. You keep ownership. DJI gets to run your footage in its marketing forever, credited, without paying you beyond any prize you happen to win.

The judging split deserves one honest note. Influence still sits on the scorecard, so reach is not irrelevant, but it is one factor of three. A small account with a genuinely original edit has a real path here, which is more than most social contests offer.

The Avata 360 Reached US Buyers Late And Through Amazon

The Avata 360 launched globally on March 26, 2026, but American pilots waited until April for stock to reach Amazon. DJI squeezed the drone through FCC certification in November 2025, 34 days before the window for new Chinese drone approvals closed for the US market.

The drone itself earned the attention. It shoots 8K/60fps HDR 360-degree video and 120MP stills from 1-inch-equivalent sensors, flies for about 23 minutes, weighs 455 grams, and carries DJI’s O4+ transmission rated to 12.4 miles (20 kilometers). Our launch coverage has the full spec rundown.

Getting it to US customers was the hard part. The Avata 360 cleared its FCC hurdle with 34 days to spare before the December 23 cutoff for new approvals. Even then, DJI’s own US store never opened orders at launch. American buyers watched Europe and Asia fly for weeks until sales opened through Amazon in April, where the Fly More Combo with the RC 2 sells for $979. That counts as a win in 2026, which tells you something.

So the contest’s premise carries a quiet catch for its American audience: to chase DJI’s prize money, you first need a drone that took a regulatory squeaker and a retail workaround to buy at all.

DroneXL’s Take

A $25,000 video contest is good marketing, and I have no problem with DJI running one. My problem is the backdrop: American pilots can win money filming a drone that reached them late, through third parties, while Washington debates whether they may buy the next one.

Yes, the security concerns behind the FCC’s Covered List push are not invented. Federal agencies have warned about Chinese drone data risks for years, and those questions deserve real answers. But a certification deadline that nearly kept a consumer camera drone out of the country, followed by a months-long retail scramble, protects incumbents more than it protects anyone’s data. The pilots paying for that are American.

The dates tell the story better than I can. Entries close September 30. The Section 232 tariff decision on drones lands September 3, right in the middle of the submission window. Whatever comes out of that decision will say more about the Avata 360’s future in this country than any leaderboard DJI publishes on October 28. Watch September 3, not the contest.

Source: DJI

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DJI Names 15 Onboard AI Challenge Winners, From Banana Farms to Bridge Cracks

Haye Kesteloo — Fri, 21 Aug 2026 17:49:00 GMT

DJI announced the 15 winners of its Enterprise Drone Onboard AI Challenge 2026 on August 20, closing out a global contest that asked developers to run AI models directly on the company’s enterprise aircraft instead of in the cloud. Judges named five Best Onboard AI Model Award winners and ten Industry Application Excellence Award winners, with projects spanning crop counting, bridge crack detection, beach litter identification, air pollution tracing, and search and rescue.

When I covered the challenge launch in March, the pitch was simple: DJI had strapped 100 trillion operations per second of computing to its aircraft and wanted to see what developers would do with it. Five months and one deadline extension later, the answer is in. It says as much about where DJI’s enterprise ecosystem is heading as it does about any single winning project.

Ten Judges Split 15 Winners Across Two Award Tiers

Ten experts from the AI and drone industries judged the entries, naming five Best Onboard AI Model winners and ten Industry Application Excellence winners. Beyond hardware prizes, winners receive placement in DJI’s Onboard AI Solutions Catalog and fast-track access to the DJI Enterprise Ecosystem partner audit.

The challenge, announced by DJI on its official contest site, covered solutions deployable on the Matrice 4 and 4D series, Dock 3, Matrice 400, and the Manifold 3 onboard computer. DJI said frontline operators are “best positioned to develop practical applications for our products,” which is why it opened its SDK and onboard computing power to outside developers in the first place, a bet I broke down when the contest opened in March.

The announcement also arrived late. DJI originally planned to name winners at the end of June, then extended the submission deadline to May 31 and pushed winners to July. The list landed August 20, weeks behind even the revised schedule.

AgroCount AI Turns a Week of Banana Counting Into Onboard Math

Daniel Tovar’s AgroCount AI, an Industry Application Excellence winner, was developed and validated at Finca La Suiza, a 124-acre (50-hectare) commercial banana plantation in Colombia. The job it automates traditionally takes four to six field workers four days, or about seven days of counting plants manually from aerial imagery.

Those labor figures come from DJI’s announcement, and independent validation isn’t public yet, but the architecture is concrete: the planned system pairs the Matrice 4E with the Manifold 3 to process imagery and GPS-tag individual plants while the drone is still flying. Growers would get field data on landing rather than a memory card full of homework. DroneXL has already documented the Matrice 4E doing serious mapping work, so a crop-counting workflow on the same airframe is a short step, not a leap.

Hangzhou New Modal Technology took a Best Onboard AI Model award for a nine-in-one fusion algorithm built on DJI FlightHub 2 and the Manifold 3. The system folds traffic enforcement, road maintenance, facility management, and traffic flow monitoring into a single closed-loop inspection workflow.

Closed-loop is the operative phrase. The system detects an issue, collects evidence, generates the alert, and supports the follow-up, all inside one workflow rather than nine separate drone missions.

The Winner List Runs Straight Through China

Twelve of the 15 winners are Chinese companies or university teams by name, from Hangzhou New Modal to Lanzhou University’s Grass Flying Team. Daniel Tovar’s Colombian project is the only winner DJI explicitly places outside China, and the announcement does not disclose how many teams entered or from where.

That entry data is the real gap in the record. DJI billed this as a global challenge, and its own press release leads with the one unmistakably international winner. (Which tells you DJI noticed the optics too.) Whether American developers entered and lost, or never entered at all, is a question DJI’s announcement leaves open, and it matters for reading what this contest actually measured.

DroneXL’s Take

The way I read this announcement, the winning projects are the least interesting part. What DJI just demonstrated is that the onboard AI era for working drones has arrived, and it is arriving almost entirely inside an ecosystem American operators are being ordered to exit. While Chinese traffic departments get closed-loop aerial inspection and a Colombian banana farm gets plant-level counting in real time, US agencies are navigating an FCC import ban that already has legacy fleets scrambling for grandfathering clarity.

The security concerns behind that ban are not invented. The Covered List proceeding is a real docket, and DJI’s data practices deserve the scrutiny they are getting. But banning the hardware does not ban the capability. It just decides who gets it, and this winner list is an early scoreboard of exactly that.

Watch the Section 232 tariff decision due September 3 for whether Washington doubles down on walling off Chinese drone hardware while this developer ecosystem compounds without American participation. If it does, the gap between what a drone can do in Hangzhou and what one is allowed to do in Houston gets wider on a schedule you can put on a calendar.

Source: DJI

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DJI Scores Its Biggest Legal Win in Years, New Drone Tariffs Hit 100%, and the HOVERAir Versa Clears the FCC

Greg Reverdiau — Fri, 21 Aug 2026 14:01:19 GMT

Welcome to your weekly UAS news update. We have three stories for you. DJI gets a big win. We have a new drone tariff. And then lastly, we have a Chinese drone maker that is somehow getting FCC clearance. So, let’s get to it.

DJI Scores Its First Big Legal Win in Years Against the Pentagon Blacklist

In the first story, DJI scored its biggest legal win in years. On August 14th, a federal appeals court partially reversed the lower court ruling in DJI’s fight against its Pentagon blacklist status. If you’re not familiar with this, DJI was added to that list back in 2022 as an alleged Chinese military company.

The court sided with the government on most points, but on one key issue, it agreed with DJI. The Pentagon claimed that DJI contributes to the Chinese defense industrial base, but that section of the report was entirely redacted. So the court said that they can’t take the government’s word for it, and now the case goes back to the lower court for another look.

DJI is still on the blacklist for now, and nothing is really changing for consumers today. But this is the biggest pushback we’ve seen yet against one of the government’s key arguments that has so far lacked public evidence.

New Drone Tariffs as High as 100% Are Coming Within Weeks

In the second story, the Trump administration just announced new tariffs on imported drones and drone parts — some tariffs as high as 100%, with drones over 55 pounds and thermal drones facing the top tariff rate. Smaller drones without those features will generally face a 25% tariff. Allied countries like Japan, Taiwan, and the EU are going to get a 15% rate, and the UK will get a 10% rate.

Most of these tariffs kick in within weeks, and the goal here is boosting US manufacturing and cutting reliance on China — which is interesting, seeing some of the other countries on the list here. However, with the FCC appearing to target foreign manufacturers’ ability to sell drones in the US at all, well, a 100% tariff on $0 is still $0.

The HOVERAir Versa Somehow Cleared the FCC While DJI Stays Blocked

And in the last story, HOVERAir just opened pre-orders for its new Versa, a pocket camera that converts into a self-flying drone — and unexpectedly, it cleared for sale in the United States. That’s right, the HOVERAir Versa is FCC approved.

That’s notable because HOVERAir’s last drone, the Aqua, was not on the FCC approval list. They missed it by weeks and had to refund over $2 million to US backers. The Versa weighs 230 grams, shoots 4K at 60 frames per second, and comes in under the FAA’s 0.55-pound registration threshold for recreational flyers.

This is kind of a stark contrast with the DJI Osmo Pocket 4, which is a camera that does not fly, and it’s still blocked from sales in the US. So it’s a bit of a strange thing, quite frankly — a flying camera gets in and the handheld doesn’t. Keeping in mind, of course, that this is a crowdfunded pre-order, not a guaranteed purchase, and shipping delays and anything else can change along the way.

The conversation is going to continue in Post-Flight, where we’re going to talk about the stories uncensored in the premium community. There’s a link in the description. We’ll see you during the live on Monday as well. Have a great weekend.

This article is based on a video from our YouTube partner Greg Reverdiau. Watch the full weekly UAS news update on the Pilot Institute YouTube channel, and find more of Greg’s work on his DroneXL author page.

DJI Ban Hits Home: Drone Pilots Ask Whether Their Businesses Survive

Haye Kesteloo — Thu, 20 Aug 2026 22:08:00 GMT

The DJI ban stopped being an abstraction in one Facebook drone community this week. Anthony Dueno, a newly certified Part 107 pilot weeks away from launching a drone services business, asked the Drone Pilots Helping Drone Pilots group whether he should quit before he starts, and the thread under his public post reads like a damage report from the bottom rung of the American drone industry. It crossed my feed at 86 reactions and 268 comments, and I have not seen the cost of Washington’s drone policy itemized this plainly anywhere else.

Dueno’s inventory of the problem runs four items: no new imports of Chinese drones, no answers on the aircraft already flying, talk of military-grade classifications, and fresh tariffs on top. “I don’t want to use anything else. I want to use DJI,” he wrote. The replies show what those four items cost in dollars.

Anthony Dueno Took His Part 107 In April And Hit The Wall By August

Anthony Dueno passed his Part 107 exam in April, formed an LLC, and flies a DJI Air 2S. Four months later, with his website nearly done, he asked a Facebook group of drone pilots whether federal policy has already killed the business he is about to launch.

Every fear in his post maps to a real policy. The import cutoff dates to December, when the FCC put every foreign-made drone on its Covered List and shut new DJI models out of the American market. The uncertainty about aircraft already flying sits in the FCC’s follow-on proceeding on previously authorized drones, the one Vic Moss and DSPA have urged pilots to file comments in before it closes September 2. And the extra 25 percent he cites on foreign-made drones is the small-aircraft rate in the Section 232 proclamation President Trump signed on August 13. For the Air 2S class he flies, Dueno has the number exactly right.

His sharpest fear is the $20,000 one: buying new equipment and ending up with aircraft that are, in his words, bricked or soon to be illegal. No agency has told him that will not happen. That silence is exactly what the open docket will resolve, and until it does, operators like Dueno are planning against a blank.

The September 3 Tariff Deadline Is Already Forcing Five-Figure Purchases

Commenters are not waiting for the tariffs to land. Several report buying thermal drones ahead of schedule because the Section 232 duties that take effect September 3 put a 100 percent rate on thermal-imaging aircraft, doubling the cost of hardware their business plans had penciled in for 2027.

Dwight Doane wrote that the deadline pushed him to buy a $7,500 thermal drone two years early, equipment he had planned for 2027 and now owns without the work to feed it. “Not happy about dropping $7500 when I don’t have it,” he wrote. He put the start date at September 2 or 3; the proclamation says September 3. The 100 percent rate he raced covers thermal-imaging aircraft and drones above 25 kilograms (55 pounds), critical components draw it as well, and smaller consumer aircraft like his pay 25 percent.

Adrian Pueyo replied in the same chain that he bought a second thermal unit for redundancy in his business. Richard Hemingway checked in 55 minutes before I captured the thread: “haha I did that today.” Three operators in one chain, pulling five-figure purchases forward to beat a duty meant to steer them toward American alternatives none of them mentioned buying.

The purchases landed the same week the White House gathered roughly 40 manufacturers for its first Drone Dominance industry event, where the agenda ran to defense orders and Pentagon lending rather than to the operators paying these duties.

Established Operators Told Him To Stay While The Thread Hunted For A Villain

The advice split along experience lines. Established operators told Dueno to keep going and find a niche. The thread then spent hundreds of replies relitigating who started the DJI ban and never settled on a villain, which mirrors the argument the industry itself has been having since 2019.

Scott Walter of Mid-Atlantic Drone Services told Dueno he started with the same equipment and that the path forward is a niche, not a panic buy: “I do think you’re panicking a little bit.” Walter himself upgraded to a DJI Matrice 4E and Matrice 4T within the last two months for infrastructure and mapping work, which is its own data point: an established operator, fully aware of the ban, still bought Chinese aircraft twice this summer because the work demands them.

The villain hunt ran longer than the advice. The thread’s most-liked comment, at 63 reactions, traces the whole campaign to Skydio’s lobbying years before either administration acted. Replies fired back that bans become real when a president signs them and pointed at the current White House and its tariffs. Others noted Congress handed the FCC its authority in the final weeks of the Biden term. (One popular theory had the administration buying DJI stock cheap after the ban tanks it. DJI is privately held. There is no stock.) The cause stayed contested all the way down the thread. The damage never was, which has been the DJI ban discourse in miniature for two years.

The argument the thread could not settle is the one the Drone Service Providers Alliance put to AUVSI in a release published the same day I captured these screenshots, pressing the trade association to put operators back at the center of its agenda and to judge aircraft on measurable risk instead of country of origin. The hardware gap underneath Dueno’s question, the absence of an American aircraft he could switch to at his budget, is the same gap DroneXL documented when the Pentagon’s $820 million component loan went entirely to military platforms.

DroneXL’s Take

My answer to Dueno is no, do not quit, and I mean that as analysis rather than a pep talk. Scott Walter built a working drone services company on the same aircraft Dueno flies today, and the demand for drone work has not gone anywhere. What changed is the cost and the uncertainty on the supply side, and both punish the people already committed hardest. That is what DroneXL has argued since the ban campaign began: this policy hurts American operators more than it will ever hurt DJI, a company that sells in every market on earth except the one closing itself off.

The forced purchases in this thread are the tariff’s first measurable effect on small business, and look at what they are: Americans spending five figures early on Chinese thermal drones. In the comment chains I read, nobody said the duty pushed them toward an American aircraft, because at $7,500 there is nothing to be pushed toward. A tariff that pulls demand forward without a domestic product to catch it works out to a tax on the transition, paid by the smallest players first.

The genuine unknown, and the question Dueno’s whole plan turns on, is whether aircraft already authorized keep flying. The FCC has proposed pulling authorizations for previously approved foreign drones and has not yet decided how far to go; the docket closing September 2 exists to settle it. Until it rules, every fleet purchase is a bet on an agency’s mood. If that bothers you as much as it bothered this thread, the comment window is still open, and Vic Moss has already written the instructions.

The White House found room for roughly 40 drone manufacturers on Tuesday. The Duenos of this industry were a Facebook thread away, itemizing the bill.

Source: Public post by Anthony Dueno in the Drone Pilots Helping Drone Pilots Facebook group, August 2026.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Navy Puts $50 Million On Shield AI's X-BAT While The Air Force Watches

Haye Kesteloo — Thu, 20 Aug 2026 20:52:00 GMT

The US Navy and the Defense Innovation Unit are putting $50 million behind Shield AI’s X-BAT, a fighter-sized jet drone built to take off and land on its tail from a trailer or a warship deck. The award, announced August 13, is the first publicly announced investment in an aircraft that has never flown. The Air Force, which Shield AI is separately courting with autonomy software for its Collaborative Combat Aircraft program, is deliberately keeping its wallet closed.

A destroyer never had a runway to lose, while the Air Force has spent two years war-gaming what happens when Chinese missiles crater its Pacific runways in a war’s opening hours. Both services see the appeal of a jet that needs neither. Only one is paying to find out if it works. DroneXL has covered Shield AI’s vertical landings since the V-BAT era, and that record says the last ten feet are where this concept gets expensive.

The Navy Wants Strike Aircraft That Fit On A Destroyer

The Department of the Navy Rapid Capabilities Office, its Portfolio Acquisition Executive for Aviation, and the Defense Innovation Unit announced the $50 million on August 13 under RIMES, a program launched in February 2026 to give surface warships a long-range strike aircraft that needs no runway or flight deck.

The award expands risk-reduction work the Navy was already running with Shield AI, per the Navy’s announcement. RIMES grew out of an arithmetic problem: a warship’s vertical-launch cells hold a finite number of missiles and cannot be reloaded at sea fast enough to sustain a long fight. The February solicitation asks for a drone that delivers 1,000-pound (454-kilogram) class munitions, the same class an F/A-18 carries, across a one-way range of at least 1,400 nautical miles (1,611 miles / 2,593 kilometers), from ships without large flight decks. The Arleigh Burke-class destroyer, the Navy’s most numerous surface combatant, can land a helicopter and nothing fixed-wing.

X-BAT is the second aircraft RIMES has backed, after the program picked Mach Industries and Whisper Aero’s hybrid-electric Atlas in June. Shield AI’s entry is the louder bet: a cranked-arrow jet the size of a fighter powered by the same General Electric F110 engine that powers an F-16, with enough thrust to climb off its tail and, in theory, settle back down on it. Shield AI cofounder Brandon Tseng framed the concept at its October 2025 unveiling: “airpower without runways is the holy grail of deterrence.”

Landing A Fighter On Its Tail Is The Hard Part

Vertical landing is where X-BAT’s design meets physics. The aircraft must descend tail-first on the thrust of its F110 engine and stay stable while its own wings catch the wind like a sail, a maneuver rockets have proven and fighter-sized drones have not.

Retired Gen. Mike Holmes, a former head of Air Combat Command, told Air & Space Forces Magazine the rocket-landing technology is understood, if not routine. Heather Penney of the Mitchell Institute for Aerospace Studies was blunter about the aerodynamics: those wings act as a sail on the way down, and “it’ll be an amazing technical feat” if X-BAT can land reliably in winds above a couple of knots.

The propeller-driven V-BAT lands the same way at a fraction of the weight, and its recovery record cuts both ways. It aced the Coast Guard’s cutter trials in 2025 and flew more than 130 sorties over Ukraine through early 2026. It also severed a Romanian naval officer’s fingers during a landing this June, and a 2024 landing mishap in front of Navy officers in Fort Stockton, Texas ended with a serious injury, Forbes reported. X-BAT weighs far more and comes down on a jet’s thrust instead of a propeller’s.

The Air Force Has Bigger Bills To Pay

The Air Force is not investing in X-BAT. Its Collaborative Combat Aircraft finalists all use runways. Analysts say the service, already paying for the F-47, B-21, and Sentinel, would rather let the Navy fund the experiment and buy the result later if it works.

“I’m not surprised that the Navy moved first,” Holmes said. Destroyers and amphibious ships already carry fuel and munitions, so X-BAT’s logistics come nearly free at sea. The Air Force version of the concept, popping jets up from hidden trailers scattered across the Philippines, would need fuel and weapons trucked to every one of those sites.

Penney’s advice to the service was to let the Navy and DIU pay, then buy the aircraft off the shelf if it proves the business case. The Air Force already has two runway-dependent CCA prototypes in flight test and the F-47 and B-21 programs consuming its budget. Shield AI puts X-BAT’s price at about $27 million, close to the Air Force’s CCA target, though that figure belongs to an aircraft that has not flown yet (price targets are never more confident than before first flight). Penney also cautioned that full mission autonomy is harder than most people realize, which puts the load on Shield AI’s Hivemind, the software that just taught Taiwan’s attack drones to fly themselves.

DroneXL’s Take

The Navy structured this the way experimental aviation should be funded: the service with the clearest use case pays to find out, and everyone else buys the answer. My worry is narrower than the concept. It is the last ten feet of every X-BAT landing.

The concession first: the V-BAT’s vertical recovery has passed real tests, including Coast Guard trials that went about as well as trials can go. But the V-BAT is a drone small enough that crews assist its landings by hand, and it has still drawn blood twice since 2024. Scale the same maneuver up to a fighter-sized jet balancing on an afterburning F110, on a moving deck, in sea wind, and Penney’s two-knot line stops sounding like snark and starts sounding like the test plan.

What I like about this structure is that nobody is pretending otherwise. DIU money exists to buy answers to exactly this kind of question cheaply, before a program of record inherits the risk. If X-BAT lands the way SpaceX boosters land, the Navy gets a reusable strike arm for 70-plus destroyers at $27 million apiece, if the price holds. If it cannot, $50 million is a cheap way to learn that, and Mach’s slower, gentler Atlas is standing right there.

Shield AI promised a first flight before the end of 2026 when it unveiled X-BAT last October. That date is now the whole story. A jet that flies in the next four months keeps the schedule credibility the price target depends on. A slip would not kill the program, but it would tell you how far the render still is from the rocket.

Sources: Air & Space Forces Magazine, US Navy, Defense Innovation Unit

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

White House Summons 40 Drone Companies As Tariff Deadline And Beijing Loom

Haye Kesteloo — Thu, 20 Aug 2026 20:08:00 GMT

The White House hosted its first Drone Dominance event with industry on Tuesday, August 18, packing roughly 40 drone companies and nearly 100 total attendees into the Eisenhower Executive Office Building. Administration officials from the National Security Council, the Department of War, and the Small Business Administration pitched a room of manufacturers on an agenda that now spans procurement, government lending, and tariffs. DroneXL has tracked this program through every phase since the February Gauntlet trials, and Tuesday marked the first time the White House itself, rather than the Pentagon, played host.

The timing did the talking. The meeting landed five days after President Trump signed Section 232 drone tariffs, two weeks before those duties take effect, and two days before Beijing demanded Washington scrap them.

The White House Ran Its Drone Push Like A Trade Fair

Roughly 40 drone companies met administration officials at the Eisenhower Executive Office Building on Tuesday for the first White House Drone Dominance event. Agencies including the Department of War, Commerce, Homeland Security, the Small Business Administration, and the FAA staffed tables companies could visit directly, Breaking Defense reported.

Speakers included National Security Council Senior Director for Counterterrorism Sebastian Gorka, Under Secretary of War for Research and Engineering Emil Michael, and Small Business Administration head Kelly Loeffler. Michael told the room the government’s job was to make sure the United States ends up “not drone sufficient, from parity, but drone dominant.”

Michael runs the program those words describe. The Pentagon’s Drone Dominance Program commits $1.1 billion across four competitive phases to buy small attack drones, and its public leaderboard shows 11 companies holding first-round deals for 24,320 aerial weapons, per Breaking Defense. A second competitive round at Fort Carson, Colorado, this month could produce orders for 60,000 platforms, the outlet reported. When DroneXL covered the first Gauntlet winners in March, the round-one commitment stood at 30,000 aircraft, so the demand signal has grown in five months.

Pentagon Lending Now Runs Alongside The Purchase Orders

The Department of War’s Office of Strategic Capital signed a conditional loan commitment of up to $820 million with Performance Drone Works on July 31 to build critical drone components in Huntsville, Alabama. At Tuesday’s event, Michael said additional deals would follow within two weeks, per the Washington Examiner.

The loan covers propulsion, power, control, and vision systems for Group 1 and Group 2 military aircraft. I examined that commitment in detail earlier this month: nothing in it is scoped to the commercial operators losing their Chinese hardware to the FCC’s Covered List, and the next names on the Office of Strategic Capital’s lending list will show whether that changes. Tuesday’s promise of more deals puts a two-week clock on the question.

Section 232 Tariffs Hit September 3 As Beijing Demands A Rollback

President Trump’s August 13 proclamation imposes tariffs of up to 100 percent on imported drones, with the top rate hitting aircraft above 25 kilograms (55 pounds), drones carrying thermal imagers, and critical components. Smaller drones pay 25 percent, and the duties take effect September 3.

DroneXL broke down the full tariff structure last week, including the 15 percent allied rate and the two Trump-linked drone stocks that rallied on the signing. On Thursday, China answered. Ministry of Commerce spokesman He Yadong said the duties would “disrupt cooperation in the global drone supply chain” and demanded Washington withdraw them immediately, the South China Morning Post reported. Beijing had already restricted its own drone exports to the United States on August 5 and blacklisted six American companies. President Xi Jinping is scheduled to visit the United States on September 24, which gives both sides a date to negotiate against.

The Administration Builds Its Case On The UFC Plot And Barksdale

Officials tied the industrial push to domestic drone threats. The FBI broke up a plot to attack the June UFC event at the White House with explosive drones, agents seized more than 600 drones at World Cup venues, and unidentified drones crossed Barksdale Air Force Base in March.

DroneXL covered each of those stories as they happened. The UFC plot allegedly paired explosive drones with a pre-staged sniper team; federal prosecutors charged five people within days, and the FBI later arrested two more. The World Cup seizures continued throughout the tournament, with the FBI’s count passing 600 by early July, the vast majority posing no threat to the crowds below. And the Barksdale incursions involved waves of 12 to 15 drones with jamming resistance and long-range control links over a nuclear bomber base, a case that remains unsolved. Gorka told the event the administration sees no way back from this technology, repeating a line he has used all month: “the future of aviation is unmanned,” per the Washington Examiner.

DroneXL’s Take

Set the room against the calendar and the event explains itself. The administration gathered 40 American drone makers inside the White House complex in the exact window between signing tariffs on their foreign competitors and those tariffs taking effect, while Beijing spent the same week cutting export flows and demanding a rollback. That is not a coincidence of scheduling. It is a government showing its domestic industry, and China, that the drone agenda now runs from the West Wing.

Some of the urgency is earned, and I will say so plainly. The UFC plot was real, Barksdale is still unexplained, and Ukraine has spent three years demonstrating what cheap unmanned aircraft do to an unprepared force. A country that builds roughly 100,000 small drones a year against China’s millions has a genuine capability problem, and gauntlet-style competition with fast orders is a more serious answer than another hearing.

My read on Tuesday, though, is that the guest list is the policy. Every dollar on display flows toward defense: attack drone orders, a component loan scoped to military aircraft, lending promises made by the Pentagon’s technology chief. The commercial operators paying the other half of this agenda’s price, the survey pilot and the fire department watching hardware costs double on September 3, were not the audience in that room, and no program announced to date is built for them. I made this argument when the $820 million loan landed, and the two weeks of deals Michael just promised will test it. If the next loan recipients build for commercial platforms, this was industrial policy. If they build only for the arsenal, American drone dominance will mean dominance of a market the government is also the only customer in, which is not what dominance meant when China won it.

Watch three dates. Comments in the FCC’s import-ban docket close September 2, the tariffs land September 3, and Xi arrives September 24. Between them sits every question this event raised and did not answer.

Sources: Breaking Defense, Washington Examiner, South China Morning Post, U.S. Department of War, The White House

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

FBI Task Force Forced Down Hobby Drones Over Spokane Wildfires

Haye Kesteloo — Thu, 20 Aug 2026 19:18:00 GMT

An FBI-led counter-drone task force used electronic countermeasures to force civilian drones out of the sky over the Spokane wildfires, after its detection gear logged 41 unauthorized aircraft inside the flight restriction in four days. The unit was assembled to protect FIFA World Cup matches in Seattle earlier this summer. It moved to Spokane County in early August when hobby pilots kept grounding helicopters and air tankers over three uncontained fires that destroyed at least 700 structures.

No pilot has been arrested. Deputies tracked operators down, cited them, and referred the cases to the Federal Aviation Administration, where Spokane County Sheriff John Nowels says each one carries up to $100,000 in fines. The authority to take the drones down is itself new: Senator Maria Cantwell said she helped secure federal approval for the task force to bring unauthorized drones down over the fires.

FBI Detection Gear Logged 41 Drones in Four Days

The FBI Counter UAS Task Force detected 26 unauthorized drones inside the Spokane flight restriction on August 2 and 13 more on August 3, reaching 41 detections in four days by August 5. At least one incursion forced aerial firefighting to stop completely while the fires burned uncontained below.

Those figures come from the Washington State Patrol’s August 5 release and a briefing by the Spokane County Sheriff’s Office. The task force draws officers from the King County Sheriff’s Office, the Spokane County Sheriff’s Office, the Seattle Police Department, the Spokane Valley Police Department, and the Washington State Patrol, and it ran drone detection around the clock inside the FAA’s temporary flight restriction, published on the agency’s TFR map, which covered the fire area up to 5,000 feet (1,524 meters) and ran 24 hours a day, in force through August 14.

On the evening of August 3, a single drone grounded every aircraft on the fires for about 30 minutes, according to the Seattle Times, and that operator was never found. DroneXL’s Rafael Suárez documented the opening days of this mess in Drone Pilots Ground Spokane Firefighting Aircraft 26 Times, when the state had 35 aircraft assigned to the fires and four drones crossed directly into their flight paths.

A World Cup Counter-Drone Unit Ended Up Over a Wildfire

The task force patrolling Spokane’s airspace was not built for wildfires. It formed ahead of the FIFA World Cup to keep drones away from packed stadiums in Seattle, moved on to the Seafair air show, and drove straight from there to Spokane County as the fires broke out.

King County Sergeant Jason Stanley described that pipeline to KIRO 7, saying the team left for Spokane before the last jet had landed at Seafair. The mission barely changed, because radar and radio frequency detection find a hobby drone over a fire line just as easily as one hovering near a stadium.

What changed is what happens next. Deputy Doug Hicks, a task force member, told reporters the unit uses electronic countermeasures against drones in the restricted airspace: “It breaks the link between the drone and the controller.” Ground crews then move in, find the pilot, and have them bring the aircraft down. Nowels told the Spokesman-Review that rumors of deputies shooting drones out of the sky were inaccurate; the tool is the jammer, not a shotgun.

DroneXL covered this unit’s stadium season as it happened, from the counter-drone school that trained officers ahead of the tournament to the more than 600 drones seized at World Cup venues. Spokane is the first fire on that resume.

Citations Went to the FAA Instead of Handcuffs

No drone pilot has been arrested over the Spokane fires. Deputies cited the operators they could locate and referred each case to the Federal Aviation Administration, with Sheriff John Nowels putting the exposure at up to $100,000 per case. Federal criminal charges remain possible on top.

An FAA spokesperson told the Spokesman-Review that hindering firefighting efforts is a federal crime punishable by up to two years in prison, and the sheriff made clear he considers these cases exactly that.

“We are done playing games with this,” Nowels said in remarks reported by KIRO 7.

These referrals are not paperwork that dies in a drawer. The operator whose DJI Mini hit a Super Scooper over the Los Angeles wildfires in January 2025 ended up pleading guilty in federal court. A Colorado pilot faced federal charges in July for grounding aircraft over the Aspen Acres Fire. As of publication, the FAA has not announced penalties in any Spokane case, so the public record on 41 detections currently ends at the referral stage.

DroneXL’s Take

The pilots who flew over these fires did more damage to this hobby than any lobbyist ever could. They handed the strongest possible case for broad counter-drone takedown authority to the people who already wanted it, wrapped in footage of their own neighbors’ burning homes.

Counter-drone authority is one of the most contested questions in American drone policy, and Congress has spent years arguing over who should get takedown powers at wildfires and under what limits. Then 41 hobby drones showed up over a burning county, and the argument settled itself in four days. A senator asked, federal officials signed off, and a jamming unit built for stadium security started breaking control links over an American city. Nobody marched in the streets about it, because nobody could defend the flights.

I get why expanded counter-UAS powers make responsible pilots nervous, and they should: authority granted in an emergency has a way of sticking around after the emergency ends. That concern is exactly why the Spokane pilots deserve the anger of every legitimate operator, with 65,000 of their neighbors under evacuation orders while they hunted for footage.

My read is that Spokane becomes the template. The model worked, and agencies talk to each other. Watch whether the FAA converts these referrals into published penalties the way it pursued the Los Angeles Super Scooper case, because that outcome is the number the next would-be fire pilot actually needs to see.

Sources: Washington State Patrol, The Spokesman-Review, KIRO 7

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DSPA Presses AUVSI To Back Operators Over Bans After Robbins Exit

Haye Kesteloo — Thu, 20 Aug 2026 15:47:00 GMT

The Drone Service Providers Alliance wants AUVSI to treat Michael Robbins’ exit as a chance to change sides. In a release published Thursday, the operator advocacy group called on the trade association to end what it describes as a reckless policy direction, put drone service providers and first responders back at the center of its agenda, and build drone cybersecurity policy on measurable risk instead of country of origin.

The release lands two days after the departure news. Robbins leaves AUVSI on September 30 to run the Semiconductor Industry Association, chief operating officer Heather Lee becomes acting president and CEO on October 1, and the board has opened a search for a permanent successor. DSPA is telling that board, in public, what kind of candidate to pick.

DroneXL has documented the distance between AUVSI’s public statements and its lobbying letters since March 2024, and DSPA’s release reads like a bill for that record coming due.

DSPA Wants Security Standards That Test The Drone, Not The Flag

DSPA’s central demand is that drone cybersecurity policy evaluate what an aircraft actually does: what data it collects, where that data goes, how it communicates, and whether any of it can be independently verified. Country of origin, the group argues, is a label on the box, not a security finding.

The release says AUVSI’s advocacy under Robbins pushed restrictions on Chinese-manufactured drones and domestic manufacturing policy without weighing the cost to the people flying drones for a living. Commercial pilots, inspectors, surveyors, educators, and public safety agencies pick aircraft on capability, reliability, availability, and price, and pulling those aircraft before affordable replacements exist lands on American businesses and on first responders mid-mission. DSPA wants standards that are rigorous, transparent, and applied to every aircraft near sensitive sites, with policy built on “measurable risk rather than geography.”

The release backs building more and better drones in the United States; what it rejects is the premise that a domestic industrial base has to be built on the backs of American operators. That is the same position DSPA’s Vic Moss took when he urged pilots to file FCC comments against the sweep that put every foreign-made drone on the Covered List in December, and it extends the group’s earlier fight over Section 2209 preemption.

A Police Drone Feed Exposure Already Proved DSPA’s Argument

DSPA argues that the person configuring a drone program, not the logo on the airframe, decides whether that program is secure. The strongest recent evidence sits in DroneXL’s own reporting: a police department’s live drone feeds spent roughly six months exposed on the open internet, and the aircraft were American-made.

DroneXL reported on July 13 that someone at the San Francisco Police Department created a Skydio ReadyLink to five live drone feeds with no access code and a one-year expiration, and security researchers Sam Curry and Maik Robert found it after about six months in the open. The exposed fleet was NDAA-compliant and built in California. Skydio CEO Adam Bry later wrote that “our product made it much too easy” to create that kind of unprotected link.

Nobody serious blamed Skydio’s country of origin for San Francisco’s configuration error, and rightly so. Swap in any manufacturer on earth and the outcome is identical. That is precisely the standard DSPA is asking Washington to apply to every drone maker, and the SFPD case shows what happens when policy obsesses over flags while the actual failure mode, a human clicking through a sharing dialog, goes unexamined.

Heather Lee Inherits The Constituency Fight DSPA Wants Settled

Lee becomes acting president and CEO on October 1, and the board has announced no end date for its search. DSPA is asking that board for a leader who treats service providers as a central constituency rather than an obstacle, and who demands evidence before endorsing new restrictions.

Robbins joined AUVSI in 2020 and has led it since 2024, a stretch in which his congressional testimony and public appearances kept returning to Chinese technology and domestic manufacturing. DroneXL reported his departure for the semiconductor lobby on Tuesday, covered him defending a Covered List sweep broader than AUVSI says it requested at XPONENTIAL Detroit in May, and published the Utah records showing the gap between the association’s webinar assurances and its lobbying letters back in March 2024. (That archive now runs past a dozen articles, which says something about how often the words and the letters diverged.)

The first test of whether a reset is real arrives before the transition does. Comments in the FCC proceeding on pulling previously authorized foreign drones close September 2, while Robbins is still in the chair, and whatever AUVSI files there will show whether the association’s direction changes with its CEO or was never his alone.

DroneXL’s Take

DSPA is right, and I’ll say so without hedging. This release puts into one document what DroneXL has argued across more than two years of reporting: the country-of-origin fight has cost American operators far more than it has cost DJI, and a logo on an airframe is not a security assessment. A secure drone earns that label through testing and verification, whoever built it. The drone industry’s trade association should have been making that case in Washington all along instead of leaving it to an operators’ alliance in Lakewood, Colorado.

The security hawks did not invent their worry from nothing. Chinese state subsidies genuinely distort the drone market, and DroneXL has opposed that distortion as consistently as it opposes protectionism dressed up as security. Drones operating near sensitive facilities deserve hard scrutiny, which DSPA itself endorses. But scrutiny is a test you run, not a flag you check, and San Francisco’s six months of exposed American-built drone feeds should end any illusion that the checkbox version keeps anyone safe.

Whether AUVSI can reset is a different question from whether it will. The association welcomed the FCC’s December sweep within hours and was still defending the broad version in May, so the burden of proof sits with the board now picking a successor. Watch the September 2 FCC filing and watch who gets the permanent job. If the next CEO’s first public sentence is about American manufacturing and the second is not about American operators, the reset DSPA is asking for died in the search committee.

Sources: Drone Service Providers Alliance release, August 20, 2026; AUVSI announcement, August 18, 2026.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

US Army Kills the Drone Battalion It Created Just Seven Months Ago

Haye Kesteloo — Thu, 20 Aug 2026 15:36:00 GMT

The U.S. Army is killing off its only drone assault battalion seven months after creating it. The 3rd Battalion, 504th Parachute Infantry Regiment, a roughly 600-soldier unit carved out of the 173rd Airborne Brigade in January to master fighting with drones and ground robots, will shed that mission after a final exercise in Germany this fall and return to regular airborne infantry duty.

The order comes from Gen. Christopher LaNeve, the acting Army chief of staff and a close ally of Defense Secretary Pete Hegseth, as part of what he calls a back-to-basics approach. The battalion was the flagship project of Gen. Randy George, the Army chief Hegseth ousted in April. The Wall Street Journal first reported the phase-out on Wednesday.

The same Pentagon is spending more than a billion dollars to put hundreds of thousands of attack drones in soldiers’ hands.

LaNeve Unwinds His Predecessor’s Signature Drone Unit

Gen. Christopher LaNeve, the acting Army chief of staff, ordered the phase-out as part of a back-to-basics push. The battalion, roughly 600 soldiers of the 173rd Airborne Brigade in Germany, keeps its drone assault role only through the Saber Junction exercise, which wraps up in September.

After Saber Junction at Hohenfels, an Army official told The Wall Street Journal, the battalion will report its findings to guide future experimentation and then “refocus on its core warfighting tasks as an airborne infantry battalion.”

The unit was activated in a January ceremony at Grafenwoehr, replacing a deactivated cavalry squadron and taking a designation that reaches back to World War II. When the Army stood it up, no end date for the drone assault role was specified.

LaNeve served as Hegseth’s senior military aide in 2025, was confirmed as Army vice chief in February, and has run the service in an acting capacity since George’s ouster. His prospects for the permanent job face resistance from senators in both parties, the Journal has reported, but the temporary status has not slowed him down. In a memo to the force on Wednesday, backed by a planning document dubbed the Army Azimuth, he wrote that soldiers must master “an operational environment that looks fundamentally different from the battlefields of yesterday.”

The drone battalion is one entry on a longer list. LaNeve is moving a Multi-Domain Task Force from Europe to South Korea, has reassigned III Armored Corps to the new Western Hemisphere Command, and clashed with Army Secretary Dan Driscoll over new fitness standards, according to the Journal’s reporting. Driscoll, for his part, has spent the year demanding better counter-drone gear for his soldiers.

The Battalion Existed to Bottle Ukraine’s Lessons

The battalion had one job: turn Ukraine’s drone war into American doctrine. Army leaders traveled to Kyiv to study it, a small number of its personnel rotated through the city, and Ukrainian troops playing the enemy in NATO training beat a full U.S. brigade combat team last spring.

George, Driscoll, and Gen. Chris Donahue, then the top Army commander in Europe, flew to Kyiv in November 2025 to meet Ukrainian drone commanders. The battalion grew out of that trip and out of George’s broader push, which he called “transformation in contact”: let the troops closest to the fight evaluate new technology and write the tactics, instead of handing those decisions to staff officers far from the battlefield.

The most telling detail in the Journal’s reporting is what happened when the two armies met in the field. Ukrainian troops played the mock adversary in a NATO training area last spring and bested a U.S. brigade combat team rotating through Europe. An army that has fought under drone-saturated skies for more than four years beat the one still writing its drone doctrine. The unit now being phased out was the Army’s answer to that loss.

Michael O’Hanlon, a defense expert at the Brookings Institution, told the Journal that George was starting to grasp how thoroughly Ukraine has changed close combat, and that “any setback in what he was doing would be deeply regretful and dangerous.”

The Army also recently canceled plans to have senior Ukrainian officers speak at its center for maneuver warfare at Fort Benning, Georgia, according to U.S. officials cited by the Journal.

The Phase-Out Cuts Against the Pentagon’s Drone Dominance Push

Washington’s official policy is drone dominance. The Pentagon is spending $1.1 billion to buy more than 300,000 small attack drones by early 2028, stood up a dedicated drone office in late June, and keeps promising unmanned systems at scale. The Army just eliminated the unit built to master them.

Hegseth has vowed to break China’s lead in drone development, with a program that reaches beyond the Army to equip the Marine Corps and Special Operations Command. His July 2025 drone dominance directive spawned the $1.1 billion Drone Dominance Program, which ran its first vendor Gauntlet at Fort Benning in February (the same post where the Ukrainian speakers were just disinvited), and this summer he created a Pentagon-wide drone czar to run it all.

An Army spokesman told the Journal the service is “committed to accelerating U.S. military drone dominance” and that these aerial systems exist to support soldiers.

What nobody has said is what happens to the mission. The spokesman declined to comment on future plans, and the Journal reports the Army did not explain how it intends to build drone warfare skills once the battalion reverts to infantry. There is a promised findings report and nothing announced beyond it: no successor unit and no timeline. An experiment can end. Ending it with nothing named to carry the mission forward is a choice.

DroneXL’s Take

Experiments are supposed to end, and that is the strongest version of LaNeve’s case. The battalion was chartered to generate findings, and one 600-soldier unit was never going to win a drone war alone. If those findings become drone platoons across the force, this story ends well.

I don’t think that is what is happening. The Journal quotes critics inside and outside the administration who see an acting chief keen to sweep away his ousted predecessor’s initiatives, and the pattern fits: George is gone, the Ukrainian officers are disinvited from Fort Benning, and the one unit whose day job was learning from Ukraine gets folded back into the infantry months after Ukrainian operators embarrassed a U.S. brigade in a mock fight. My read is that the drone mission is collateral damage in a changing of the guard, and back to basics is the branding.

When the Army was pushing 340,000 drones toward its combat units at Fort Irwin in March, the hardware sprint and the tactics experiment looked like two halves of one plan. Now the Pentagon is buying the drones and disbanding the tacticians. Hardware without doctrine is how you end up with 300,000 aircraft and nobody who has thought hard about employing them against an enemy that has.

Watch what surfaces after Saber Junction wraps in September. If the battalion’s findings show up as funded drone formations in the Army’s force design, I will happily be wrong. If they show up as a report on a shelf, remember that the soldiers best qualified to write the next one just went back to being riflemen.

Source: The Wall Street Journal

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Northcom Cannot Stop A Homeland Drone Swarm. Its Deputy Commander Said So.

Haye Kesteloo — Wed, 19 Aug 2026 15:34:00 GMT

Asked at a defense symposium in Huntsville on Thursday, August 13, whether he could defend the United States against a drone swarm launched from inside its own borders, the deputy commander of U.S. Northern Command answered in one word. No.

Army Lt. Gen. Joseph Jarrard, who also serves as vice commander of the U.S. element of NORAD, did not soften it afterward. He said the command lacks the sensors to see a swarm forming and, depending on where it went, may lack anything to shoot back with.

I have been reporting the American counter-drone buildout piece by piece since the fiscal 2026 defense bill passed in December: the police mitigation authority, the FEMA money, the FBI schoolhouse in Alabama, the fly-away kits. Jarrard’s answer is what all of that adds up to nine months later.

Jarrard Named Detection As The Constraint, Not Authority Or Money

Jarrard told the Space and Missile Defense Symposium that Northern Command lacks both the sensors to find a swarm and the effectors to stop one. Detection is the harder half. His command is now trying to tie military radars into systems local police already operate.

“We are ill-equipped right now to handle something like that,” he said, in remarks first reported by The War Zone, which attended the symposium. He pointed to Operation Noble Eagle, the post-9/11 air defense posture, as the doctrinal template his command is now adapting for counter-drone work, and to the World Cup as the working model for federal coordination with local law enforcement.

That framing matters more than it sounds. Northern Command was designated the Pentagon’s lead synchronizer for homeland counter-drone operations in November 2024. Twenty-one months later its deputy commander is describing interoperability with municipal police radars as the path forward, which is an admission about how thin the military-only picture is. DefenseScoop reported that Jarrard is the second Northcom general in four months to say a version of this out loud.

Barksdale Showed The Gap Is Wider Than Sensors

Security forces at Barksdale Air Force Base counted multiple waves of 12 to 15 drones over the Louisiana flight line between March 9 and 15. The sensors worked well enough to count them. Five months on, no one has publicly identified who was flying.

A confidential March 15 briefing document obtained by ABC News described aircraft with non-commercial signal characteristics, long-range control links, and resistance to jamming. DroneXL reported that campaign in March, over a base that hosts B-52 bombers and nuclear weapons storage. The investigation is still open.

The detection story is genuinely improving. Gen. Gregory Guillot told Congress in March that Northcom now defeats about a quarter of the drones it detects over installations, against almost none a year earlier. He also disclosed that the command had exactly one deployable counter-drone fly-away kit at the time, crewed by 11 soldiers. A twenty-five percent defeat rate against a single mobile kit is not a defense. It is a pilot program with a good press release.

The Offense Crossed Its Own Line Nine Days Before Jarrard Spoke

Nine days before Jarrard spoke, Bloomberg watched five quadcopters lift off a Maryland cornfield and attack a Cold War-era tank without a pilot flying any of them. The software running that swarm came from Auterion, and it already ships to Ukraine by the tens of thousands.

Katrina Manson’s August 14 account for Bloomberg describes an operator touching a controller once on August 4, after which the drones organized themselves into formation and ran the engagement. Cut their link to the operator and they keep talking to each other and finish on preloaded targeting.

None of this is a prototype. Auterion’s Nemyx engine ran the first multi-manufacturer kill chain in December, and a single operator destroyed three separate targets with it at Camp Blanding, Florida on January 8 under the Pentagon’s own Swarm Forge program. The hardware is deliberately interchangeable. Any aircraft running the company’s stack joins a swarm through a software update.

Meanwhile the rules governing that software are being rewritten on a clock. NSPM-11, signed June 5, gives the Secretary of War 90 days to issue an updated DoD Directive 3000.09 on autonomy in weapon systems. That lands on or about September 3.

The Pentagon Is Buying Faster Than It Can Cover Ground

The fiscal 2027 request seeks $14.4 billion for counter-drone defenses at more than 250 sites. The Defense Department tracks 824 installations. Congressional Budget Office analysts put comprehensive layered defense of a single benchmark base at roughly $74 million, plus $5 million a year to run.

Run the CBO arithmetic and 100 protected sites cost about $7.4 billion up front, as The Center Square reported from the budget documents and the agency’s July analysis. The contracts are moving regardless. Joint Interagency Task Force 401 signed CACI to a $500 million SkyValor deal in July, AeroVironment took a separate $500 million award, and the Marine Corps put $11 million into Epirus this month for a vehicle-mounted high-power microwave system built off the Leonidas that fried a 49-drone swarm in Indiana last summer.

Assistant Secretary of Defense for Critical Technologies Michael Dodd put the department’s appetite plainly on August 12. “Bring us your best. Not slideware,” he wrote, asking industry for lighter airborne directed energy weapons. Directed energy is also where the domestic program has already gone wrong once: a laser fired near the border without proper coordination shut down El Paso airspace in February, and the FAA did not clear military anti-drone lasers for U.S. airspace until April.

DroneXL’s Take

Jarrard deserves credit before he gets criticism. When drones flew over Langley for 17 straight days in December 2023, the answer from the Pentagon was silence and then euphemism. This time a three-star said “no” into a microphone in front of the defense industry. That is a better failure mode than the one we had.

The problem is what the “no” is actually about. Read the coverage and you come away thinking the United States needs more sensors. Barksdale says otherwise. The sensors there worked. Somebody counted the drones, wave by wave, logged their signal characteristics, noted that they resisted jamming, and wrote it into a classified briefing five days in. What did not happen was attribution, and what did not happen was interdiction. Detection was never the binding constraint at the one incident we have the best documentation for.

My read is that $14.4 billion buys hardware for a problem the department can define and skips the one it cannot. Sensors and effectors are procurable. Knowing who launched from a farm road eight miles outside the fence, and having the legal posture to do something in the ninety seconds before impact, is not a line item. The FY2027 request covers 250-plus sites out of 824 and nobody has explained which 574 are acceptable losses. That is the question I would put to Jarrard, and it is not the one he was asked.

I made the domestic-policy half of this argument yesterday, about Texas, so I will keep it to one line here: every cycle like this one ends with new rules, and new rules get enforced against the pilots who broadcast Remote ID rather than the ones who do not.

The date to watch is on or about September 3, when the 3000.09 rewrite NSPM-11 ordered comes due. The demonstration in Maryland put the lethal decision in a software layer that the current directive still defines through the aircraft carrying it. If the update leaves that definition alone, the United States will spend the next budget cycle buying microwaves to shoot down machines its own policy no longer knows how to classify.

Sources: The War Zone, DefenseScoop, Bloomberg, National Defense Magazine, The Center Square, Air & Space Forces Magazine, White House NSPM-11.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Germany Opens Counter-Drone Test Center After Leipzig Explosive Drone

Haye Kesteloo — Wed, 19 Aug 2026 15:03:00 GMT

Germany opened a national research center for drone security on Tuesday, exactly two weeks after a bomb-carrying drone struck the wing of a Ukrainian Antonov An-124 at Leipzig/Halle Airport. Interior Minister Alexander Dobrindt, Research Minister Dorothee Bär and Saxony-Anhalt premier Sven Schulze inaugurated the Drone Security Technology Center at the German Aerospace Center’s Cochstedt airfield, about 25 miles (40 kilometers) south of Magdeburg and roughly 60 miles (100 kilometers) from the apron where that drone lay undetected for four hours.

The center will develop and test systems for detecting and defeating hostile drones at a working airport, with a second site at the DLR campus in Braunschweig. Dobrindt called the threat part of the “shadow war of the 21st century” and announced that Germany’s federal police counter-drone force will double.

DroneXL flagged this opening on August 10, when Dobrindt told Bild am Sonntag that the drone technology arms race demanded research units of Germany’s own. Tuesday delivered the building.

DLR Puts Counter-Drone Research on a Working Airfield

The Drone Security Technology Center operates at two DLR sites, Magdeburg-Cochstedt Airport and Braunschweig, where researchers will develop and field-test systems for detecting and defeating hostile drones under real airport conditions. DLR will invest more than 10 million euros in the buildout and add up to 40 new positions.

Cochstedt started as a Soviet military airfield and later limped along as a regional airport before DLR bought it in 2019. The agency’s National Test Center for Unmanned Aircraft Systems went operational there in 2021, and the airport resumed flight operations a year later. Utilization has climbed about 20 percent annually since, passing 200 usage days in 2025 with roughly half of those booked by outside companies and research partners, according to DLR’s announcement.

Drones at Magdeburg-Cochstedt Airport, where DLR tests how drone traffic and conventional air traffic control share the same airspace (archive image). Photo credit: DLR (CC BY-NC-ND 3.0)

The new center bolts a security mission onto that infrastructure. Manufacturers from startups to established players can bring detection and interception systems to the field for testing, and the center’s work feeds into the Joint Drone Defense Center, where Germany’s security agencies pool their drone picture. DLR chair Anke Kaysser-Pyzalla framed the mission as transfer speed: “Research must have an impact, especially where the security of our country is concerned” (translated from the German announcement).

Dobrindt said the facility is unique nationwide and sits deliberately at airport infrastructure, the exact setting where the threat has landed.

The Leipzig Drone Bomb Set the Timeline

On August 4, a quadcopter carrying Semtex and PETN crossed the Leipzig/Halle Airport fence and struck the wing of a parked Ukrainian An-124. The charge tore loose and lay on the apron for about four hours before a bus driver spotted it. Only a failed detonator kept the outcome forensic.

DroneXL has tracked the case since the first report on August 6. Investigators confirmed the charge held military-class explosive and assessed the drone was built to slip past the airport’s installed defense system. By August 10, DNA on the drone had matched a trace from the 2024 arson attack on the airport’s DHL hub, with the same profile on file in Lithuania, and federal police had stood up an EchoShield radar, built by Echodyne of Kirkland, Washington, on the airport’s old control tower.

The wider pressure is documented. Dpa counts 166 unauthorized drones over Germany’s 15 international airports in the first half of 2026, and suspected reconnaissance drones shut Munich Airport twice in one week last October.

Germany’s federal prosecutor general has taken over the Leipzig case. U.S. intelligence assessments reported by the Wall Street Journal link the drone to the Russian government, while German authorities have made no attribution and Moscow denies involvement. That gap is the unresolved center of Tuesday’s ceremony: Berlin opened a test range for defenses against an attacker it will not yet name.

Berlin Doubles Its Drone Police While the Research Arm Spins Up

The Cochstedt center is the research layer of a defense buildout that already includes hardware, manpower and law. Germany’s federal police counter-drone force is doubling from 150 to 300 officers and from four to eight bases, and police have held shoot-down authority since October 2025.

The doubling, which Dobrindt confirmed at the opening, matches what he told Bild am Sonntag on August 10. Lawmakers gave federal police explicit authority to disable or shoot down threatening drones ten months ago, and a national drone defense center in Erding has been running since May. What Germany lacked was a place to find out which systems actually work before buying them. Leipzig made that gap concrete: the airport had a defense system installed, and the attacker engineered around it.

Bär pointed to IDAS-PRO, a research project aimed at spotting and capturing rogue drones around airports and large events, and promised on Instagram that unwanted drones over stadiums or concerts would get a “not with us.” The defense ministry sounded different. Spokesperson Mitko Müller warned after the Leipzig discovery that “you cannot provide detection and defence capabilities around the clock” along tens of thousands of kilometers of fencing, and that the notion of one hundred percent security is deceptive. (One of those two will age better.)

DroneXL’s Take

When I closed our August 10 report, I wrote that Cochstedt’s opening would show whether Berlin’s answer to Leipzig is capability or another coordination body. Eight days later I can answer it: this one is capability. A live airfield, controlled airspace, an incubation track for manufacturers, and a direct line into the police units that will fly whatever passes the tests. Testing a jammer in a lab tells you it works in a lab. Cochstedt can tell you whether it works between cargo flights.

The concession is owed here: none of this touches a legitimate pilot, and none of it should. The Leipzig device carried military explosives into a NATO cargo hub. Counter-drone research aimed at that threat is the legitimate safety work DroneXL has always backed, and it is the opposite of grounding hobbyists after every scare. An attacker willing to strap Semtex to a quadcopter ignores a geofence either way.

My worry is scale. More than 10 million euros and up to 40 positions is seed money against what Dobrindt himself calls an arms race, and the escalation since last autumn only bends one way. Copenhagen and Oslo closed in September 2025, and Munich followed twice in October. Ten months later the mystery sightings had graduated to a charge on an Antonov’s wing.

Two things to watch. The Vilnius parcel trial is in session, and if the Leipzig DNA match survives scrutiny it drags the drone case into that courtroom. And DLR says the Cochstedt buildout runs across the next several years, so the race that matters is whether tested hardware reaches those 300 officers before the next drone crosses a fence. Müller’s arithmetic about tens of thousands of kilometers of fencing says Berlin already knows it cannot win that race everywhere. Cochstedt exists to make sure it wins where it counts.

Sources: DLR, Euronews

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DJI Flip Hits $349 on Amazon Two Weeks Before Trump's 25% Drone Tariff

Haye Kesteloo — Wed, 19 Aug 2026 14:54:00 GMT

The DJI Flip is selling for $349 on Amazon again, $90 under its $439 list price, and this round of the recurring discount runs into a hard deadline: President Trump’s new drone tariffs take effect September 3, with consumer aircraft like the Flip in the 25% tier.

All four Flip configurations carry a cut as of Wednesday, August 19. The RC-N3 bundle is $349, the RC 2 screen-controller bundle is $509, down from $639, the Fly More Combo is $619, and a two-battery combo without a controller is $309. DroneXL has covered this drone since its January 2025 launch and flown it through a six-month long-term test, and the verdict has been the same throughout: this is the drone to hand a first-time pilot. What’s new in August is the deadline.

Get the DJI Flip (RC-N3) for $349 on Amazon

The DJI Flip takes off from an open palm, no landing pad required. Photo credit: Air Photography

Full Propeller Guards Make the Flip a Drone You Can Fly Near People

The DJI Flip is the first DJI drone built with full-coverage propeller guards, folding carbon fiber cages that surround all four props. The guards let the drone launch from an open palm, land back in your hand, and fly closer to people than any Mini should.

The guards are why the Flip exists. Folded, they collapse with the arms into a package that fits a jacket pocket. Unfolded, they let a nervous beginner take off indoors, hand-launch from a boat, or run AI subject tracking at a kid’s birthday party without anyone flinching. Forward 3D infrared sensing brakes the aircraft automatically when it detects an obstacle ahead, though coverage is not omnidirectional, so a branch coming in from the side still wins.

The one honest weakness is wind. In Air Photography’s six-month long-term review on DroneXL, Shawn found the guards act like sails when you fly back against a stiff breeze at altitude, and that finding has held up since. The Flip is composed at distance and height in calm air. If the trees are moving, stay low or pick a different drone.

The Camera Is the Same 1/1.3-Inch Sensor DJI Put in the Mini 4 Pro

Behind the guards sits a 1/1.3-inch CMOS sensor with an f/1.7 lens, shooting 48-megapixel stills, 4K HDR video at 60 frames per second, 4K slow motion at up to 100fps, and 10-bit D-Log M color, all stabilized by a three-axis mechanical gimbal.

That sensor is the reason DroneXL called the Flip ugly in May and told you to buy it anyway. The DJI Mini 4 Pro carries the same chip in a $759 aircraft, and footage from the two is hard to tell apart. The 10-bit D-Log M profile gives you grading latitude that palm drones like the DJI Neo 2 don’t offer, and stills save as RAW (DNG) files.

Each battery is rated for up to 31 minutes, and DJI’s O4 transmission holds a video link to 8.3 miles (13 kilometers) under FCC rules. Because the aircraft stays under 249 grams, recreational pilots in the United States skip FAA registration and Remote ID. Every other rule still applies at 248 grams (registration is the paperwork you skip, not the responsibility).

The RC 2 Bundle at $509 Buys a Screen, Not a Better Aircraft

Both discounted bundles contain the identical aircraft. The $349 version pairs the Flip with the RC-N3 controller, which uses your phone as the display. The $509 version swaps in the RC 2, a controller with a built-in 5.5-inch, 700-nit screen.

Buy the $349 RC-N3 bundle if flying off your phone screen doesn’t bother you; most first-time buyers should stop right there. Step up to the RC 2 bundle at $509 if you want the built-in display and a controller that works on its own, with nothing to mount and nothing to cable up.

The RC-N3 (left) uses your phone as a display; the RC 2 (right) brings its own screen. Photo credit: Air Photography

Two other configurations are discounted alongside them. The Fly More Combo with the RC 2 and three batteries is $619, down from $779, and a two-battery combo without any controller is $309, though that last listing was shipping from a third-party seller rather than Amazon directly at the time of writing, so check the seller box before checkout. The no-controller combo only makes sense if you already own an RC-N3 or RC 2; older DJI remotes from earlier Mini models won’t pair with the Flip.

Trump’s 25% Drone Tariff Takes Effect September 3

President Trump signed Section 232 tariffs on imported drones on August 13. Consumer aircraft like the Flip fall in the 25% tier, which starts September 3, 2026. The duty applies to declared import value rather than retail price, so the shelf impact depends on what importers pass through.

DroneXL broke down the full proclamation when the president signed it: thermal and heavy aircraft pay 100%, docking stations and listed components are covered, and nearly everything American hobbyists fly lands in the 25% bracket, per the White House proclamation.

The tariff arrives on top of an FCC blockade that is already in place. The agency added foreign-made drones to its Covered List on December 22, blocking authorizations for new DJI models, which makes the approved aircraft already on Amazon’s shelves most of what American buyers get to pick from. The Flip has cycled down to $349 before, on Black Friday and again in early July, and each time the price recovered. Whether it cycles back after September 3 is a different bet, because from that date every replacement unit crossing the border carries the new duty.

DroneXL’s Take

The $349 is not the news. This drone has touched that number at least three times since Black Friday, and left to itself the sale would come and go again the way sales do. September 3 is what changed. Fifteen days from now, a 25% duty attaches to the declared import value of every consumer drone entering the country, and the honest answer on what that does to shelf prices is that nobody outside the importers knows: the duty hits customs value, not the sticker, and Commerce still hasn’t published the criteria for its onshoring relief program. What the record shows is that when stacked duties briefly pushed the burden on Chinese drones toward 170% in April 2025, Vic Moss of the Drone Service Providers Alliance warned prices would double for small operators, and plenty of pass-through showed up fast.

So my position is simple. If a Flip was already on your list, buy it before September 3, because whatever version of this deal exists on September 4 gets priced off tariffed inventory. If it wasn’t on your list, the tariff is a bad reason to put it there; a deadline sells drones the way it sells anything, and the Flip earns a purchase on its camera and its guards or not at all. Watch what Amazon’s DJI listings do in the first week of September. That is when we find out whether 25% at the border becomes 25% at checkout.

Source: Amazon

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Amazon Prime Air Promises 500 Cities This Year, One 15-Mile Circle At A Time

Haye Kesteloo — Wed, 19 Aug 2026 13:47:00 GMT

Amazon plans to offer Prime Air drone delivery in nearly 500 US cities and towns by the end of 2026, a sixfold increase over its current footprint, the company announced Wednesday. New launch sites in the Chicago, Atlanta, Cleveland, Syracuse, and Boise metro areas will join the 11 markets where Prime Air already flies across seven states.

That headline number deserves a second look before anyone pictures 500 drone ports. Each Prime Air site covers roughly 175 square miles (453 square kilometers), and Amazon counts every municipality inside that circle toward the total. Standing next to the MK30 delivery drone at XPONENTIAL Europe in Düsseldorf this April, what struck me was how much aircraft delivers how little payload. The same ratio runs through this announcement: the fleet is modest, and the arithmetic is generous.

Amazon Counts Every Town Inside Each 175-Square-Mile Circle

Prime Air currently flies from 11 sites in seven states, and each site serves an area of roughly 175 square miles (453 square kilometers). Amazon’s “nearly 500 cities and towns” tally counts every municipality those circles touch, so a handful of new metro launches multiplies the number quickly.

Work Amazon’s own multiplier backward and today’s 11 sites already count as roughly 80 cities and towns, a bit over seven municipalities per circle. Drop the same circle on the south suburbs of Chicago, where Prime Air plans to fly from fulfillment centers in Markham and Matteson, and the count climbs far faster, because 175 square miles is a circle roughly 15 miles (24 kilometers) across, and suburban Cook County packs dozens of incorporated villages into that much ground.

The five named metros put Prime Air in Georgia, Ohio, Illinois, Idaho, and New York for the first time, with what Wednesday’s announcement calls more communities to come later this year. At least one launch carries a caveat Amazon’s own team supplied: in May, Prime Air infrastructure principal Jeff Cleland told Syracuse residents that the Clay, New York operation hoped to fly this year but could slip into 2027 depending on FAA approval timelines. Syracuse still sits on today’s list.

The urgency has a name, and it’s Walmart. About 90 percent of Americans live within 10 miles (16 kilometers) of a Walmart store, and Walmart’s partner Wing is under contract to fly from more than 270 of them by 2027, an advantage DroneXL mapped in July. Amazon has to draw each circle around a warehouse instead. A sixfold jump in the municipality count is how a warehouse network answers a store map.

Prime Air Halved Its Fees And Widened The Catalog Before Scaling Up

Drone delivery is now free for Prime members on orders of $50 or more. Smaller Prime orders cost $2.99 and non-members pay $4.99. Nearly any item under 5 pounds (2.3 kilograms) that fits a large shoebox qualifies, which Amazon says covers more than 60% of frequently purchased products.

The pricing is the quiet story here. When Waco launched last November, Prime members paid $4.99 per drone delivery and everyone else paid $9.99. Nine months later the same service costs $2.99 and $4.99, with the free tier on top. Companies don’t halve fees on a product they’re winding down.

Amazon's shopping app now surfaces drone delivery at checkout, letting Prime Air customers pay $4.99 for one-hour delivery and pin the exact backyard drop-off spot on a satellite map. Photo credit: Amazon

The volume figures remain Amazon’s own. “We’ve already delivered hundreds of thousands of packages to customers by drone this year,” said David Carbon, vice president of Prime Air. CNBC reported that Carbon told employees in a March all-hands that the division expects to reach 1 million deliveries in 2026, and CEO Andy Jassy’s annual shareholder letter set targets of communities with 30 million customers by year-end and half a billion packages annually by 2030. Most orders arrive around 60 minutes after checkout, with the fastest at about 30. “It’s still an experiment,” Forrester retail analyst Sucharita Kodali told the Associated Press.

The MK30’s Incident Record Travels With The Expansion

Five new metro areas will inherit an aircraft with a documented incident history: a two-month nationwide grounding in early 2025, two crane strikes in Arizona, a severed internet cable in Waco, and an apartment-building collision in Richardson, Texas. No serious injuries have occurred across any of these events.

Amazon voluntarily paused all US commercial drone operations in January 2025 after two MK30 drones crashed at its Oregon test site, where a software update had made lidar altitude sensors read light rain as the ground. The FAA approved a fix that March and flights resumed. The incidents that followed through late 2025 and early 2026 all involved stationary obstacles, a record DroneXL compiled when Amazon pitched Central New York on its 176-square-mile delivery zone.

Amazon’s answer is its Detect-and-Avoid system, which the company takes seriously enough that it quit the Commercial Drone Alliance in March over whether such systems should be mandatory across the national airspace. In more than 70,000 flights, Amazon says, the system twice steered its drones away from crewed aircraft, one of them a helicopter that wasn’t broadcasting its position. The company holds Part 135 air carrier certification, the same framework used for commercial airlines, along with FAA waivers to fly beyond visual line of sight, and its 11th market opened in Papillion, Nebraska, just three weeks ago. The drones fly suburbs, not downtowns, and steer clear of major airports.

DroneXL’s Take

The 500-city figure is a counting choice, not a construction plan. Amazon measures municipalities its delivery circles overlap, and one suburban circle can sweep in dozens. The plan is real, the growth is real, and the headline number still flatters the map more than the fleet.

Concede what’s real first: this is no longer the program that limped through 2024. Fees got cut in half, thousands of packages move daily, and the January 2025 grounding produced an FAA-approved fix instead of a quiet shutdown. Thirteen years after Jeff Bezos told 60 Minutes that drone delivery was four or five years away, packages actually arrive by air in 11 American markets. That deserves to be said plainly.

What I don’t trust is the unit of measurement. The announcement never says how many launch sites the 500-town map requires, and the site count is the number that determines capital spending, jobs, airspace load, and how many planning boards Amazon has to win over. Syracuse shows why that missing number matters: Amazon’s own infrastructure principal allowed in May that Clay could slip to 2027, yet Syracuse appears on today’s list of metros launching soon. Winter isn’t a press release variable either. The MK30 stays on the ground below 14 degrees Fahrenheit (minus 10 Celsius), which puts real holes in a Boise, Cleveland, or Central New York December, exactly the month the year-end count comes due.

My scorecard for this promise is short. Watch whether Clay and Markham actually fly before the FAA’s environmental assessments and local site-plan approvals run out of calendar, and watch the December accounting: if “nearly 500” survives it, Amazon will have earned the number the same way it defined it, one 175-square-mile circle at a time. Half a billion packages a year by 2030 works out to well over a million drone flights a day. Today’s operation flies thousands. Between those two figures sits every hard question this announcement doesn’t answer.

Sources: Amazon, CNBC, Associated Press

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DJI Sets September 1 For SDR Transmission 2. US Buyers Are Blocked.

Haye Kesteloo — Wed, 19 Aug 2026 00:00:00 GMT

DJI has set a launch event for Tuesday, September 1 at 20:00 GMT+8, which lands at 8 a.m. Eastern, and the teaser shows a small metal box with two fold-up antennas lit from above. Retail packaging photographed by a leaker gives it a name. DJI SDR Transmission 2 Combo.

The company’s Taiwan account posted the teaser with the Chinese tagline 小有實力, 大陣從容, translated as “Small in strength, big in composure.” In the video version, the chassis carries a red RX button beside a blue TX button, with an HDMI port along the top edge. That points to one unit that switches between transmitting and receiving, instead of the separate transmitter and receiver DJI has sold since 2024.

DroneXL covered the first SDR Transmission teaser cycle in July 2024, five days ahead of launch, down to the FCC identifier already sitting in the agency’s database. That filing is the piece that cannot happen again. The Federal Communications Commission added DJI’s communications equipment to its Covered List in December 2025, and listed equipment cannot receive the authorization a new model needs to be imported or sold in the United States.

Leaked Packaging Names A TX/RX Switchable Combo

Leaker Igor Bogdanov photographed a retail box reading DJI SDR Transmission 2 Combo, with TX/RX SWITCHABLE printed across the bottom in English and Chinese. Two identical units sit on the front, each with fold-up antennas and a square status display. DJI has confirmed none of it.

DJI’s own teaser video, posted by its Taiwan account, gives a date and a tagline and nothing else. Bogdanov published the box shot separately, and he has a long record with DJI packaging leaks, including the Osmo Pocket 4 retail box and the Neo 2.

Packaging for the DJI SDR Transmission 2 Combo shows the ribbed, TX/RX-switchable transmitter and receiver units with fold-out antennas and small status screens. Photo credit: DJI

The display render on the right-hand unit reads 4K 60fps HDR. The current SDR Transmission tops out at 1080p60, so if that readout survives into shipping hardware it is the headline change, and a packaging mockup is exactly the kind of artwork that gets finalized before firmware does. One body doing both jobs also changes the buying math. Today a crew needs a transmitter and a receiver as separate SKUs, and a spare of the wrong one is a paperweight.

DJI teases a new product launch set for September 1, 2026 at 20:00 GMT+8, showing only the silhouette of a compact, antenna-equipped device. Photo credit: DJI

DJI Cleared The First SDR Transmission Through The FCC Before Launch

An FCC filing for DJI’s original SDR transmitter, ID 2ANDR-TX53209, was submitted by SZ DJI Osmo Technology Co., Ltd. on June 14, 2024. The product launched roughly five weeks later, on July 17, at $549 for the transmitter and receiver together.

Those numbers are worth holding onto. The 2024 system sends 1080p60 at up to 20 Mbps with 35 ms of latency once camera and screen are excluded, reaches 10,000 feet (3 kilometers) under FCC power limits, and weighs 145 grams with antennas. Standalone units ran $309 each. It was built for small crews who could not spend thousands on a Teradek-class link.

The regulatory path was ordinary then. DJI filed, the lab tested, the grant issued, the product shipped, and we wrote it up from the leaks before DJI said a word. The Lito series was among the last to run that sequence, reaching an April 2026 launch on filings submitted before the December deadline closed.

The Covered List Reaches DJI Radios, Not Only DJI Drones

The FCC’s December 22, 2025 action did two things. It added all foreign-produced drones and drone critical components to the Covered List. For DJI and Autel specifically, it went further and covered all communications and video surveillance equipment from both companies.

That second half is what catches a cinema video link. An SDR transmitter is a radio that moves a picture from a camera to a monitor, which reads as communications equipment on its face. DJI argues otherwise. Its petition for reconsideration, which the Commission’s procedural notice records as filed on January 21, 2026, asks the FCC to confirm that the listing cannot reach products falling outside the statutory definition of equipment essential to advanced communications service, a category DJI says excludes most of its catalog. The company put $1.56 billion on that table in April, roughly $860 million of it tied to twenty-five models slated for release in 2026.

Whether a 145-gram monitor link qualifies as equipment essential to advanced communications service is the precise question DJI put to the Commission in January, and the Commission has not answered it. The reconsideration record closed to replies on May 11. DJI’s parallel Ninth Circuit petition, case 26-1029, sits in abeyance waiting for that same answer, with its opening brief due November 2.

Three Federal Deadlines Land On Consecutive Days In September

DJI’s event falls on September 1. The FCC’s comment window on Public Notice DA 26-758, which proposes pulling import and marketing clearance from foreign drones the agency already approved, closes September 2. Section 232 tariffs on imported drones and components take effect September 3.

The tariff annexes are written around unmanned aircraft and their components, and no DJI cinema accessory is named in them. The Covered List already does that work on the DJI-specific side. The White House proclamation signed August 13 sets 100% on drones above 55 pounds (25 kilograms) and on any drone carrying a thermal imager, with 25% on nearly everything else American pilots fly.

American crews keep what they already own. The 2024 SDR Transmission was authorized before the listing, so remaining US stock stays legal to sell and buy, and the FCC has waived the firmware cutoff for pre-listing equipment through at least January 1, 2029. New models are the casualty, and the only route left for a determined buyer is a grey-market import with no US warranty and no support. Competitors are moving into that space with the paperwork DJI cannot get, and HOVERAir’s VERSA cleared the FCC for US sale this month while DJI stayed blocked.

DroneXL’s Take

This is what the ban looks like once it stops being about drones. A 145-gram box that sends a picture to a director’s monitor is not an airspace threat and not a surveillance platform. It got swept up because DJI made it and it has an antenna. American cinematographers on small productions were the people the first SDR Transmission was priced for, and they are the people who lose a $549 option they can no longer legally buy.

The security case is not empty, and I want to be fair about that. The Department of War told the FCC in April that it participated in the interagency body behind the determination and transmitted a classified annex to Congress, which is a real answer to DJI’s claim that no qualifying agency ever acted.

It arrived on April 3. The listing took effect December 22, and the statute gave the department thirty days from the determination to file that report. Ten weeks late, months after the deprivation, is not a security judgment supporting a decision. It is a decision looking for one. I think the calendar drove December 22, and everything filed since has been an effort to build the file that should have existed first.

If the Ninth Circuit reaches the merits on scope when briefing opens November 2, rather than sending the whole thing back to the Commission, the question of whether a video transmitter counts as equipment essential to advanced communications service gets a judicial answer instead of an agency one, and that answer reaches every DJI product with a radio inside it. Until then, DJI will keep launching hardware on Shenzhen time that Americans can watch and cannot buy.

Sources: DJI Taiwan, Igor Bogdanov (@Quadro_News), FCC Public Notice DA 26-223, FCC Fact Sheet, December 22, 2025, Center for Agricultural Law and Taxation, White House Fact Sheet, August 13, 2026, CineD, DJI

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Duke University Drone Tracks Another Drone Using Only Its Sound

Haye Kesteloo — Tue, 18 Aug 2026 22:43:00 GMT

A quadcopter at Duke University can find and follow another quadcopter using nothing but the sound its rotors make. No GPS sharing between the two aircraft, no radio link, no camera, no ground infrastructure.

The system is called SonicFly, and a team at Duke’s General Robotics Lab posted it to arXiv on August 1, 2026, with a project page and flight footage following two days later. Across outdoor pursuit runs in darkness, fog, direct sun and overcast, the follower held a target separation of 11.5 feet (3.5 meters) with a mean distance error of 4.4 feet (1.34 meters). DroneXL has been covering acoustic drone detection since early 2024, and this is the first working system we have seen that puts the microphones on an aircraft that is itself flying and making noise.

The array weighs 10 grams.

SonicFly Separates The Leader’s Rotor Noise From The Follower’s Own

The hard part is not hearing a drone. It is hearing one while flying another. Duke’s follower sits inside its own acoustic field, so the team gave the two aircraft different propellers: two blades on the leader, three on the follower. That harmonic gap is what makes the leader audible.

Multirotor sound is not featureless hiss. Blade passage produces a fundamental tone set by blade count times motor RPM divided by 60, with harmonics stacked above it, so a listener knows roughly where in the spectrum to look. The follower carries four TDK InvenSense MEMS microphones on an 83.5 mm square board with 75 mm spacing, sampling at 12 kHz. A Jetson Orin Nano turns those channels into spectrograms plus inter-microphone phase and level differences, runs a convolutional network that outputs the leader’s bearing and range, gates out low-confidence estimates, and filters the rest.

The useful band turned out to be narrow. The network did best between 1 and 2 kHz, and feeding it the full 0 to 6 kHz range made results worse. Low frequencies carry the most rotor energy, but their long wavelengths give a 75 mm array almost nothing to work with directionally. Higher frequencies resolve direction more finely, then fade with distance and start wrapping phase.

Onboard alternatives to satellite positioning are a running thread here, from vision systems that map their own way through GPS-denied airspace to swarms built to hold formation when the radio link dies. Sound is the channel a radio jammer cannot reach.

The Acoustic Payload Weighs 10 Grams And Draws One Watt

Size, weight and power is where SonicFly makes its clearest case. The microphone board occupies 128 cubic centimeters, weighs 10 grams and draws 1 watt. Duke measured an Intel RealSense D455 at 75 grams and 3.46 watts, and a Livox Mid-360 LiDAR at 265 grams and 6.5 watts.

Both aircraft are Holybro X500 V2 quadrotors on Pixhawk 6X controllers with RTK positioning. The RTK data never reached the perception stack: it labeled the training set and scored the results afterward, and the estimator saw none of it. The follower got BLHeli_32 speed controllers and 900 kV motors to haul the extra load. The 10-gram figure covers the microphone board, not the Jetson bolted on beside it.

The Tracking Numbers Show A Proof Of Concept, Not A Product

Bearing is where the system struggles. After confidence gating and filtering, mean absolute bearing error was 31.2 degrees, improved from 45.7 degrees raw. Range held up better at 1.49 meters RMSE. The follower stayed inside the desired lateral corridor for 78% of total flight time across trials.

Training and testing ran at separations of 4.9 to 23 feet (1.5 to 7 meters), the follower flying about 20 inches above the leader to stay out of its wake. Error did not fall off smoothly with distance. It was lowest at intermediate separations, because up close the leader’s near-field wake scrambles the signal, and far out the harmonics sink into the follower’s own broadband roar.

The contrast with stationary listening is stark. Twelve ground units spread across a 98 by 66 foot (30 by 20 meter) field resolved bearing to 6.86 degrees and range to 1.26 meters, roughly five times better than the flying array managed. That gap is the whole difficulty of the problem in one number. The Duke authors are direct about it, describing a system that demonstrates pursuit and relative-state maintenance rather than precise formation control. This is also an arXiv preprint rather than peer-reviewed work, and the figures come from the team that built the thing.

Acoustics Outperformed Cameras And LiDAR In Degraded Conditions

Duke scored all three modalities on the same flight data, gathered in evening light and haze, counting the share of time each could locate the leader within 5 meters. Acoustics managed 76.57%. LiDAR reached 61.13%. Vision came last at 52.48%. Fog and darkness cost a microphone nothing.

The flight campaign spanned 46 to 90 degrees Fahrenheit (8 to 32 Celsius), humidity from 19% to 93%, and winds up to 23 mph (37 km/h), all of which bend and scatter sound in ways a lab bench never shows. Leader paths included straight lines, C-curves, L-turns and, because graduate students are graduate students, a route that spelled out “DUKE.”

Ground-based acoustic detection is already a commercial category rather than a laboratory curiosity, from the cheap Ukrainian microphone arrays that spawned North American imitators to the sensor networks strung across Ukrainian towns to call Shahed inbounds and Belarusian systems on the same principle. What none of those do is fly.

DroneXL’s Take

This story is going to get read as counter-drone technology, and that read is premature. A follower that tracks a leader by ear sounds like an interceptor, right up until you notice that the Duke team engineered the target to be audible in the first place.

Two-blade propellers on the leader, three on the follower, picked so their harmonics would separate. A hostile drone will not accommodate you like that, and the paper says plainly that this work covers a single leader-follower pair. Whether the same network can hear an aircraft it was never trained on, one whose blade count nobody selected for contrast, is not answered anywhere in the results.

I read the real result differently. The interesting number is 10 grams and 1 watt for a sensor that works in fog and darkness without line of sight, at a fraction of what a LiDAR puck costs in payload. That is not a weapon. That is redundancy for aircraft flying in formation, and formation is what falls apart first when GPS goes away and the radio link gets stepped on. Ukraine has spent three years proving how routine both conditions have become.

The counterpoint deserves a hearing, because ground-based acoustic detection has earned its place: cheap microphone arrays are doing real counter-drone work today, and a listener that ignores weather and daylight is obviously attractive to that market. Fine. But 31.2 degrees of bearing error inside a 23-foot envelope is not a detection system, and dressing it up as one does the researchers a disservice.

The version that matters is the one where the blade-count crutch comes out and a follower tracks an unmodified consumer quadcopter it has never heard before. Publish that, and the counter-drone read becomes fair. Until then what stands is that a flying robot can navigate by the noise its neighbor makes, which is enough. Universities keep producing the drone research that moves this field, as Houston’s work on formally provable flight behavior showed last month, and the industry keeps skipping the limitations section to write the press release.

Sources: Duke University General Robotics Lab, arXiv:2608.00401, IEEE Spectrum

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

AUVSI's Robbins Exits For SIA As The Drone Ban He Backed Rolls On

Haye Kesteloo — Tue, 18 Aug 2026 22:29:00 GMT

Michael Robbins will leave the Association for Uncrewed Vehicle Systems International on September 30 and take over as president and CEO of the Semiconductor Industry Association on October 1. AUVSI named chief operating officer Heather Lee acting president and CEO effective the same day, and its board has opened a search for a permanent successor.

Robbins spent three years telling American drone operators that their industry needed what the chip industry already had. He is now going to work for the chip industry. DroneXL has been covering him since March 2024, when Utah public records showed AUVSI putting written support for an immediate purchase ban on Chinese drones in front of a state senator while its own president told a webinar audience the association opposed exactly that.

Robbins asked Washington for a CHIPS Act for drones and got the ban half instead

The semiconductor comparison was Robbins’s own. In a February 2024 AUVSI webinar he said the association wanted federal industrial policy for drones modeled on what Congress did for chips. “We would like to see a chips act for drones,” he told the audience.

That package never arrived. What arrived instead was the restriction side of the same playbook: NDAA Section 1709, the Section 1260H designation of DJI, Section 805 procurement cutoffs, more than a dozen state bills, and on December 22, 2025, the FCC sweeping every foreign-made drone and critical component onto its Covered List. American operators got the tariff-wall economics without the factory-building money. Robbins is leaving for the industry that received both.

AUVSI counts roughly 7,500 members worldwide. Even if every one of them flew in the United States, that would be under two percent of American Part 107 certificate holders, a point DroneXL made in 2024 and one worth restating now that the association’s advocacy has reshaped what the other 98 percent are allowed to buy.

AUVSI’s public position on Chinese drones never matched its lobbying letters

AUVSI’s public statements and its letters to state legislators say different things, and the paperwork is public. DroneXL published the Utah documents in March 2024, the Law Enforcement Drone Association rebutted Robbins in writing a year later, and the pattern repeated in California weeks after that.

On February 27, 2024, Robbins told an AUVSI webinar that nothing the association supported would ban Chinese drones for service providers or public safety, and that any suggestion otherwise was false. A Government Records Access and Management Act request in Utah surfaced a January 23, 2024 email from an AUVSI government affairs manager to Senator Wayne Harper expressing support for Part 12 of SB 135, the section prohibiting purchases of drones made or assembled by a covered foreign entity. That is a ban. It was submitted in writing five weeks before the webinar.

LEDA president Jon Beal answered a Robbins op-ed in February 2025 by accusing AUVSI of “overt gaslighting” on Chinese drone legislation, and DroneXL covered that rebuttal in full. Six weeks later the association’s regulatory counsel sent California lawmakers a letter opposing AB 1160, which DroneXL took apart line by line: contested labels, no dates, no breaches, no dollar figures. Pilot Institute and 51 Drones pulled out of AUVSI’s annual conference over this stretch of advocacy.

Robbins welcomed the FCC ban, then defended a version broader than AUVSI says it asked for

Robbins welcomed the FCC’s December 2025 Covered List action within hours. His statement argued the United States had to build its own drones instead of depending on Chinese supply. Five months later he defended a policy his own association says it never requested.

Opening XPONENTIAL Detroit on May 12, 2026, Robbins acknowledged AUVSI had lobbied for a narrow, risk-based Covered List implementation and that the White House went far wider. He then made the affirmative case for the wider version, citing reindustrialization goals and pre-conflict industrial-base contestation. DroneXL was in the room. DJI, whose Ninth Circuit petition against the FCC was already pending, was not on the stage to answer.

By February 2026 the same approach had been ported to a new sector. AUVSI launched a Partnership for Robotics Competitiveness on Capitol Hill using language nearly identical to its 2023 drone campaign, with Unitree, AgiBot, and UBTech standing in for DJI. DroneXL documented the reuse at the time.

Heather Lee inherits an open FCC docket and a drone registration database for sale

Lee, AUVSI’s chief operating officer, becomes acting president and CEO on October 1, and the board has announced no end date for its search. She arrives with comments closing September 2 on the FCC’s proposal to pull previously authorized foreign drones and DJI’s Ninth Circuit challenge still undecided.

She also inherits a product DroneXL reported on three weeks ago. AUVSI is selling subscription access to a drone registration intelligence database built from more than one million U.S. registrations, priced between $5,000 and $8,500, with a stated use case that includes tracking foreign equipment penetration by manufacturer. The FAA itself stopped publishing that bulk data. Every operator who bought a DJI fleet is now a data point AUVSI can sell, and the buyers include the companies competing for that operator’s next purchase.

Robbins said in SIA’s announcement that it is a privilege to join the association and represent an industry foundational to American economic strength. AMD chair and CEO Lisa Su, this year’s SIA board chair, said Robbins brings the vision and experience the semiconductor industry needs. Neither statement mentions drones.

DroneXL’s Take

Give Robbins this much: he opposed Florida’s immediate ban, he opposed the Countering CCP Drones Act as written, and he pushed for transition periods when plenty of lawmakers wanted none. Those were real positions and they helped real public safety agencies. I have never argued otherwise.

The problem is what the transition-period position turned into. Vic Moss and Kyle Nordfors both told Robbins to his face on that 2024 webinar that a transition period is a ban with a delay fuse, because no agency spends money on a fleet it will have to replace. They were right. By December 2025 the FCC had banned new foreign drones outright, by May 2026 Robbins was on a Detroit stage defending a sweep broader than the one AUVSI says it lobbied for, and by August 2026 he has a better job at the association whose industrial policy he held up as the model. American drone operators are paying higher prices for less capable hardware. The CHIPS Act for drones is still not a bill.

My read is that this exit is a promotion earned on the drone file, and the drone file is not finished. Watch who AUVSI names permanent CEO and what the association files in the FCC’s previously-authorized-equipment proceeding this fall. The board has announced a search and no timeline, and AUVSI has not said whether its next leader will keep defending a Covered List broader than the one it claims to have wanted. That silence is the thing to track, because the position was never voted on by the operators it affected.

Sources: Semiconductor Industry Association announcement, August 18, 2026; AUVSI announcement of Heather Lee as acting president and CEO, August 18, 2026; Robbins’s post on X; AUVSI webinar transcript published by DroneXL, February 27, 2024.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Helsing Pitches Japan A Strike Drone With A Contested Combat Record

Haye Kesteloo — Tue, 18 Aug 2026 21:16:00 GMT

Rakuten is brokering the sale of German strike drones to Japan’s army. The Tokyo group confirmed on Monday that it is helping Munich-based Helsing finalize a deal to supply unmanned systems to Japan’s Ground Self-Defense Force, which is field-testing the company’s HX-2 loitering munition through the end of September.

That puts one of Japan’s best-known consumer brands inside a weapons procurement, and it puts a European strike drone in front of a military that has spent this year trying to stand up its own. DroneXL reported in July that Helsing had picked West Virginia for its first American factory, building the same aircraft and pitching it to the Pentagon.

Neither company said how many drones are in play or what Japan would pay.

Rakuten Steps Into Arms Brokering Without A Defense Track Record

Rakuten confirmed on Monday that it is helping Helsing close a sale to Japan’s Ground Self-Defense Force. The Tokyo company runs online retail and a mobile carrier. It has no weapons business of its own, and Japan’s Nikkei newspaper reported its role in the deal before either side spoke publicly.

Helsing said it reached an initial agreement with the Japanese government earlier this year, and credited an unnamed local brokerage partner with making that possible. Hideaki Mukai, chief of staff to Rakuten CEO Hiroshi Mikitani, told Reuters that his company wants to widen the door for defense startups, which he said face “significant structural barriers, including complex procurement systems” in Japan.

This is not Rakuten’s first pass at the job. It arranged an April demonstration for a Japanese military unit by Swarmer, the US-listed Ukrainian company whose software flies drones in coordinated groups, and it has helped Ukrainian defense firms exhibit at Japanese arms fairs. Mikitani has been one of Japan’s loudest corporate backers of Kyiv. A company most Americans know for cash-back shopping links is now the front door to Japan’s army.

Ukraine’s Frontline Tests Left The HX-2 Record In Dispute

Bloomberg reported in January that Ukraine had paused further HX-2 orders after frontline trials by its 14th Regiment. An internal German Defense Ministry presentation described takeoff failures and absent autonomy features. Helsing rejected that account and said Ukrainian units had asked for more than 1,000 additional aircraft.

The presentation, dated November 20, 2025 and prepared by a group reporting to General Günter Schneider, said the drones lacked components meant to deliver terminal guidance, midcourse navigation and visual target acquisition. Bloomberg also reported that jamming near the front cut the link between the aircraft and their operators, and that Berlin would not fund follow-on orders unless Kyiv asked for them. A German Defense Ministry spokeswoman said her management had never reviewed or approved the document.

Helsing’s response was specific. The company said it had no knowledge of the presentation, denied a high failure rate on launch, and said the regiment that ran the trials had since requested more than a thousand more drones and that the HX-2 sits in Ukraine’s central military ordering system. “The hit rate of the first flights, which was officially documented, is encouraging,” the company said, disputing the reporting as based on an unapproved internal file.

The aircraft itself is an electrically propelled X-wing munition of about 26 pounds (12 kilograms), quoted at 137 mph (220 km/h) and a reach of 62 miles (100 kilometers). Helsing unveiled it in December 2024 as Europe’s answer to Russia’s Lancet.

Japan Is Shopping Abroad While Building Its Own Drone Industry

Japan approved a record 9.04 trillion yen ($58 billion) defense budget for the fiscal year that began in April, and named uncrewed systems one of its spending priorities. Its acquisition agency picked four domestic drone makers out of 38 applicants this month. Foreign suppliers are circling the same customer.

Tokyo’s cabinet scrapped the five noncombat categories that had limited defense transfers on April 21, clearing Japanese industry to sell any category of finished weapon abroad rather than only the noncombat ones. Reuters reported that the rule change, stacked on top of the spending, has pulled overseas defense firms toward Japan hunting local partners. Anduril is in talks to buy Nissan’s shuttered Oppama plant near Yokosuka to build drones there, and opened a Japanese unit in December.

All of that lands on a country that is simultaneously trying to build the capability itself. The four finalists in the Acquisition, Technology and Logistics Agency’s rapid development program went to navy testing in early August, under a Takaichi government that treats coastal defense by uncrewed aircraft as its answer to Chinese carriers operating closer to Japanese waters every year. Buying German while building Japanese is not a contradiction. It is what a government does when it thinks the clock is short.

The Same Drone Is Headed For A West Virginia Production Line

Helsing chose Martinsburg, West Virginia, in July for its first US plant, a $50 million site designed to turn out more than 2,000 HX-2 drones a month. The company targets initial operating capability in November and full-rate production within a year. It says it is negotiating with the American military.

Dozens of Helsing HX-2 strike drones sit ready for shipment atop stacked transport cases inside the company's production facility. Photo credit: Helsing

The Martinsburg factory is meant to create at least 60 jobs at an average salary near $125,000, and US general manager Jennifer McArdle told WV MetroNews last month that “we expect the US military to be procuring our product.” Helsing says the US Army has already tested the aircraft during exercise Flytrap in Lithuania. The company closed a $1.8 billion round in July at an $18 billion valuation, which makes it the best-funded defense technology startup on the continent.

DroneXL’s Take

Give Helsing its due first, because the criticism is easier than the record. The HX-2 is in a shooting war, thousands have gone to Ukraine, Helsing keeps engineers on the front line, and the drone is listed in Kyiv’s central ordering system. A single internal slide deck that the ministry which produced it says its own leadership never approved is thin ground for calling a weapon a failure.

My problem is the evidence base. Every public number about how well the HX-2 actually works comes from Helsing or from that disowned document. Neither Kyiv nor Berlin has released independent launch or hit-rate data, and Helsing’s Flytrap testing claim is Helsing’s own. Japan has not said what threshold the drone has to clear, or whether it will publish anything at all when the tests end. Rakuten, whose expertise is retail logistics and mobile networks, is not the party that closes that gap.

None of that makes the HX-2 a bad buy. It makes it an unaudited one, and unaudited is how a lot of drone procurement is being done right now, in Tokyo and in Washington both. The Pentagon is about to be pitched this same aircraft off a Martinsburg line that opens in November, and it will be pitched on the same combat reputation Japan is testing this month.

Which is why the end of September matters more than Monday’s announcement. Japan’s field tests close then. If Tokyo publishes what it found, allied buyers get the first independent read on this aircraft that anyone outside Helsing has offered. If Tokyo says nothing and signs anyway, the HX-2’s combat record stays exactly what it is today: a vendor’s word against an unapproved slide deck, on a weapon two more governments are about to buy.

I would rather Japan tell us. So would the American taxpayer footing the next order.

Sources: Reuters, Bloomberg, The Defense Post, Helsing, WV MetroNews, The Diplomat

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

HOVERAir VERSA FCC Certification Was Granted In Error, FCC Says

Haye Kesteloo — Tue, 18 Aug 2026 21:14:00 GMT

Update, August 21: This article was published August 18 under the headline “HOVERAir VERSA Clears FCC For US Sale While DJI Stays Blocked,” based on HOVERAir’s own statement that the VERSA held FCC certification and on an equipment authorization that appeared in the FCC database under the company’s grantee code. The FCC has since told DroneXL that the certification was granted in error, has been set aside, and has been removed from the public database. The article has been rewritten. The agency’s full statement appears below.

The FCC has ordered HOVERAir to stop importing, marketing and selling the VERSA in the United States, and says the equipment authorization the company pointed to when it opened Indiegogo pre-orders on August 18 should never have been issued.

In a statement provided to DroneXL, an FCC spokesperson said the certification was granted erroneously, has been set aside, and has been pulled from the agency’s public database. The spokesperson said the Commission is weighing further action against HOVERAir for what it described as an apparently willful violation of the Covered List rules.

The reversal lands four days into a crowdfunding campaign that has already cleared its funding goal, and it leaves backers who paid on the strength of that certification holding a device the agency says cannot lawfully be sold to them.

The FCC Says Both Halves Of The VERSA Are Covered

The agency’s position is that neither configuration escapes the Covered List. The camera by itself counts as a prohibited UAS critical component produced in a foreign country, and once the Flight Kit is attached the product becomes a foreign-produced UAS. The full statement, attributable to an FCC spokesperson:

“Importing, marketing, or selling this device in the United States is a violation of U.S. law, and the FCC has directed HoverAir to cease any such activity immediately. The device in question is on the FCC’s Covered List: the Versa camera alone is a prohibited UAS critical component produced in a foreign country and, when the Flight Kit is included, the device falls under the ban on UAS produced in a foreign country. The certification for this device was granted erroneously and has been set aside by the FCC and removed from our public database. The FCC is considering further action against HoverAir for apparently willfully violating the FCC’s Covered List rules. Finally, the FCC is sending guidance to all Telecommunications Certification Bodies to strengthen their due diligence of applications to ensure they comply with their legal obligation not to certify equipment on the FCC’s Covered List.”

The last sentence reaches well past this one company. Telecommunications Certification Bodies are the private labs the FCC accredits to issue equipment authorizations on its behalf, and the agency is telling every one of them that a grant slipped through that should have been refused.

ZeroZero Sought Certification For The Camera Alone

The workaround is now documented. ZeroZero Robotics applied for authorization covering the camera system rather than the complete aircraft, as The Verge established this week, and its marketing never uses the words drone or UAS. The spec sheet calls the VERSA an intelligent self-flying camera.

Asked about the ban, ZeroZero marketing director Anita Facundo told The Verge the product was “in full compliance with the applicable regulatory requirements” for the configuration sold to US customers. Gizmodo reported the same framing, noting the propeller cages were treated as an accessory rather than as part of an aircraft.

The FCC has now rejected that reading in the plainest terms available to it. A camera built to dock into a flight module is a UAS critical component whether or not the module ships in the same box.

HOVERAir Has A Second Campaign Facing Refunds

DroneXL covered the VERSA reveal on August 4, when the company had not said whether it would seek US authorization at all. The answer turned out to be yes, through a route the FCC now says was never available to it in the first place.

This is the second HOVERAir crowdfunding campaign in twelve months to collide with the same rule. The waterproof AQUA launched globally in May with the US excluded after the company failed to secure authorization before the FCC added all foreign-produced drones and their critical components to the Covered List on December 22, 2025. More than 1,800 backers who had put in over $2 million were offered refunds instead of hardware.

US backers of the VERSA campaign are now in the position AQUA backers occupied last winter, with one difference that cuts against them. The AQUA never had a certification to lose. The VERSA had one, and the agency took it back.

The Specs Still Stand Even If The Sale Does Not

HOVERAir lists 230 grams, 4K video at 60 frames per second in 10-bit HDR, a 3-axis mechanical gimbal, up to 14 minutes of flight, and a top speed of 22 mph (36 km/h). None of that is in dispute, and none of it is what the FCC objected to.

The HOVERAir Versa pairs a handheld camera grip with a caged self-flying drone body, shown here in white and black. Photo credit: HOVERAir

The Verge reported a 1/1.3-inch sensor. Level 5 wind resistance tops out around 24 mph (39 km/h), the same rating 51 Drones found workable but gust-sensitive on the X1 PRO MAX across seven months of testing. At 230 grams the VERSA also sits under the FAA’s 250-gram registration threshold, which was always a separate question from the FCC one and remains unaffected by this week’s action.

The Indiegogo campaign opened worldwide on August 18 and passed its funding goal within a day. HOVERAir had pointed to a fall 2026 retail release and a press preview on September 3 ahead of IFA Berlin.

DroneXL’s Take

I got this one wrong on Tuesday, and the correction matters more than the original story did. I published a piece arguing the Covered List screens for filing dates rather than security risk, and I used the VERSA’s certification as the evidence for it.

The certification was a mistake by a certification body, not a policy outcome. My argument rested on a fact that was not a fact, and I owe readers that plainly rather than buried in an update line.

What I will not walk back is the underlying objection. Look at what the FCC has now confirmed: a private lab issued a grant it should have refused, nobody at the agency caught it until a Chinese manufacturer had banked crowdfunding money from American backers, and the fix is a memo asking labs to try harder. That is not a screen. A rule enforced by press statement three days after the fact, on a product that four major outlets had already told readers was legal to buy, is a rule that is not working as designed. American backers are the ones eating the loss, again, and they are the same people locked out when DJI’s Osmo Pocket 4P launched globally on July 30.

HOVERAir does not come out of this clean either. The company built a product that separates into two pieces, sought certification for one piece, and marketed the pair. Calling the propeller cage an accessory was a legal theory, and the company tested it with other people’s money. The FCC’s word for that is willful.

The guidance to Telecommunications Certification Bodies is the part to watch. If the agency follows through, the next few months should produce more set-aside grants, because it is unlikely the VERSA is the only device that got through this way. Whether any of that reaches HOVERAir as a fine, and whether backers see refunds before the campaign closes, are the two open questions I would put to the company today.

Sources: FCC spokesperson statement provided to DroneXL, The Verge, Gizmodo, HOVERAir, Indiegogo

Correction: An earlier version of this article, published August 18, reported that the HOVERAir VERSA held valid FCC certification for sale in the United States. The FCC has since told DroneXL that the certification was granted in error and has been set aside. The article has been rewritten and retitled.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Texas Wants Counter-Drone Power. Only 61 US Officers Are Certified.

Haye Kesteloo — Tue, 18 Aug 2026 21:12:00 GMT

The Texas Senate Select Committee on Homeland and Border Security heard on August 12 that Mexican cartels are flying the same drones American police fly, and that 61 officers in the entire United States are certified to bring one down.

El Paso Police Sgt. Cory Balke told the committee his department has tracked 1,776 cross-border drone incursions. He asked lawmakers for more authority at the local level, and for the training and money to use it. One of those requests has already been answered, and not by Austin.

I have spent the past year tracking the federal counter-drone buildout for DroneXL, from the first FEMA grant tranche through the FBI certification school running out of seats in the middle of the World Cup. What Texas senators heard last week is that same bottleneck, described from the Rio Grande instead of from a House hearing room.

El Paso Police Have Tracked 1,776 Cross-Border Drone Incursions

Balke told the August 12 hearing his department has logged 1,776 drone crossings from Ciudad Juárez, a count limited to aircraft broadcasting Remote ID. Radar returns and sensor hits over the same period put the number closer to 14,000. Those two figures rest on very different kinds of evidence.

The 1,776 count covers aircraft carrying what Balke called digital license plates, the Remote ID signal that commercially manufactured drones broadcast by design. A radar return identifies nothing. It registers an object. How many of the roughly 14,000 were drones, and how many were birds or light aircraft, is the difference between a serious problem and an overwhelming one, and the testimony as reported does not settle it. Balke made the sharper point himself: anything crossing without a broadcast signal never enters the tracked column at all.

On capability he was blunt. Cartel operators across the river fly $30,000 aircraft that can identify a person at a mile (1.6 kilometers) and follow a moving vehicle, and the inventory runs from $500 consumer models up to $60,000 six-rotor agricultural machines able to haul 441 pounds (200 kilograms) of cargo in a single trip. “The threat is very high, and they are very capable,” Balke said of the operators in Juárez. Balke told senators that running a drone program is the natural foundation for a counter-drone program, and his own department is the working example: El Paso police fly a citywide drone-as-first-responder network built on 22 DJI Dock stations across nearly 40 miles (64 kilometers) of urban and border airspace, which DroneXL covered last August.

Congress Already Granted the Mitigation Authority Texas Is Asking About

The SAFER SKIES Act, signed in December 2025 as part of the fiscal 2026 defense authorization, extended counter-drone mitigation authority to state, local, tribal and territorial agencies. A Justice Department and Homeland Security interim final rule made that authority operational on July 1, exactly six weeks before the hearing.

The law amends 6 U.S.C. 124n, the statute that had reserved drone mitigation to a short list of federal agencies since 2018. Under the interim final rule, certified state and local officers can disrupt, seize, disable or destroy an aircraft that meets a credible threat standard, using only equipment on a federal authorized systems list. Unauthorized mitigation carries civil penalties reaching $100,000 per violation. DroneXL covered the act’s reach over stadiums and amusement parks in March.

A state legislature cannot hand its officers the power to jam or capture an aircraft. The prohibitions sit in the Communications Act and the federal aircraft sabotage statutes, and only Congress can lift them, as the Congressional Research Service lays out. The committee’s own interim charge concedes the arrangement: it directs members to determine which state agencies would best assist the federal government in drone mitigation, not which ones Texas can deputize by itself. Chair Phil King framed the hearing around whether “current state laws and enforcement authorities are sufficient.” On mitigation, state law was never the operative document.

The Certification Pipeline Still Runs Through One FBI Schoolhouse

Every officer cleared to take down a drone under the new rule trains at a single facility, the FBI’s National Counter-UAS Training Center at Redstone Arsenal in Huntsville, Alabama. As of the July rule, roughly 46 agencies and 61 individual officers nationwide had finished mitigation certification. Sixty-one.

That total lines up almost exactly with the spring cohort. DroneXL reported in May that the bureau was training about 60 state and local officers ahead of the World Cup, and the July rule counted 61 certified. By June, Homeland Security Secretary Markwayne Mullin was telling House appropriators that the school had more applicants than seats while the tournament was still running. The certification is also individual rather than institutional: one trained officer does not authorize a department.

The two tiers are not equally hard to clear. Detection and warning certification is a free online course, and the departments expect around 1,500 agencies to take it. Mitigation is a two-week residential course at one campus in Alabama. The tier that lets an officer actually stop an aircraft is the one stuck at double digits, which is why a Lawfare analysis in June argued for pushing certification out to state peace officer training commissions instead of running everything through one federal pipeline. Money is not the binding constraint either. FEMA’s counter-drone grant program is already moving hundreds of millions of dollars, and Texas DPS has been expanding its own drone fleet for years.

The Ukraine Casualty Figure Cited in Austin Comes From Kyiv

Drone technology specialist Michael McGee told senators that drones caused more than 90% of Russian combat losses in March, roughly 34,000 troops. That figure originates with Ukraine’s own military, which counted 35,351 March casualties and credited 96% to unmanned systems, making it a combatant’s claim rather than an independent count.

The magnitude holds up better than the attribution. CSIS estimated in July that Russian casualties ran between 30,000 and 34,000 per month during 2026, which brackets the number McGee used. Ukraine’s defense ministry said every strike in the tally carries video confirmation, though the 96% drone share remains Kyiv’s own arithmetic. McGee’s underlying warning is better supported than his statistics: DroneXL documented cartels copying Ukrainian drone tactics back in November 2025, and the transfer of battlefield technique to the border is observable rather than theoretical.

His framing deserves more scrutiny than it got. McGee, a former fighter pilot who holds a doctorate, told the committee that August 12 was the equivalent of August 12, 2001, and that “our next 9/11 is coming” in 30 days. Thirty days from the hearing lands on September 11, 2026, the 25th anniversary. That is calendar rhetoric, delivered to a committee that was weighing appropriations. No threat assessment sits behind the date.

DroneXL’s Take

The cartel drone threat is real, and I want that on the record before I criticize the response. This is not New Jersey in December 2024. Balke’s operators exist, their hardware is documented, and DroneXL has tracked cartels adopting Ukrainian tactics for the better part of a year.

When Texas law enforcement says the people across the river are as capable as they are, I believe them.

The problem is that Texas convened a hearing to solve a problem Texas does not own. Congress already granted the mitigation authority in December. The Justice Department and DHS already built the operational framework in July. What is missing is throughput at one schoolhouse in Huntsville, and no bill filed in Austin can widen that door. My read is that the legislature’s honest options are narrow and unglamorous: fund seats at NCUTC, fund detection gear that clears the authorized systems list, fund the DFR programs that Balke correctly identified as the foundation, and lobby the federal delegation to decentralize certification.

What worries me is the other half of the committee’s charge. It directs members to identify necessary prohibitions and strengthen penalties, and that is the part a state legislature actually can do quickly. A hearing about $60,000 cartel aircraft carrying 441 pounds across the Rio Grande has a way of producing statutes that land on the Part 107 operator photographing a construction site in Houston. The cartel pilot in Ciudad Juárez has never once checked a Texas penal code provision before launching. Every new prohibition written this interim will be enforced almost exclusively against people who did.

Two dates are worth tracking. Public comment on the DOJ and DHS interim final rule closes September 6, which is where any serious push to move certification into state training commissions has to appear on the record. The committee’s recommendations then land before the legislature convenes in 2027. If those recommendations arrive heavy on new state penalties and light on funding for federal certification seats, Texas will have spent an interim legislating against the wrong pilot.

Sources: Bethany Blankley for The Center Square (syndicated copy), Texas Senate Select Committee on Homeland and Border Security, Border Report, Federal Register, Congressional Research Service.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Hillsborough County Hasn't Voted on Its Blood Drone Program Yet

Haye Kesteloo — Tue, 18 Aug 2026 21:11:00 GMT

Hillsborough County commissioners vote on September 2 on whether to approve the whole blood drone delivery program that Tampa General Hospital, Hillsborough County Fire Rescue and Archer First Response Systems announced at the start of this month. Until that vote lands, nothing flies. The October start date the partners keep citing is conditional on it, and no public document yet says what the county would pay.

That contingency has been dropping out of the coverage as the story travels. Trade write-ups published as late as August 17 put the word “launches” in the headline. DroneXL covered the original announcement on August 4 and carried the vote then, and we have been following this particular company since September 2020, when it was still pitching drone-delivered medical gear to a neighboring Florida county and waiting on the FAA.

Hillsborough County Has Not Approved the Program

The Hillsborough County Board of County Commissioners takes up the whole blood drone program on September 2. Tampa General Hospital’s own announcement states the technology is scheduled to come online in October 2026 subject to that approval. Without the county’s authorization, the program does not exist.

That language sits in the August 3 press release from Tampa General, the document that started the cycle. It is the final paragraph, after every quote in the release, which is roughly where a communications team puts something it would rather not lead with.

Two ArcherFRS systems go in at Fire Rescue Station 9 near Faulkenburg Road and Sabal Park, and at Station 37 in Riverview, together covering about 70 square miles (181 square km) of the county with the heaviest volume of penetrating trauma. A 911 call flagging severe bleeding launches the drone alongside the ambulance. ArcherFRS puts delivery at under three minutes on average, a company projection with no independent field data behind it yet, since the system has not flown a single Hillsborough call. Our August 4 report covered the clinical case for putting whole blood in the field.

The Company Press Release Dropped the Contingency

Archer First Response Systems issued its own release on August 7, four days after Tampa General’s, describing the program in the present tense throughout. That version omits the county vote and the October contingency entirely. Tampa General’s announcement, which came first, states both.

Read the ArcherFRS release knowing a funding vote is pending and the drafting gives itself away. It talks about agency budgets facing real uncertainty, and about getting tools to the field even when near-term funding paths are not fully settled. Rescue Division Chief Todd Carnell appears calling the system “a true force multiplier in the hands of our crews”, a quote that reads very differently before a budget vote than after one.

The same release went out a second time on August 12 through a different wire service, and that version quietly drops OneBlood, the blood supplier, from the partner list. Neither release explains the change.

Manatee County Built the Template and Exposed the Weak Point

Manatee County ran the first ArcherFRS deployment in 2024 under a contract costing the county one dollar a month, with Florida Health Sciences, the entity operating Tampa General, covering most of the expense. ArcherFRS now brands that structure the Champion Hospital Partner Program.

That rate covered the initial one-year term. A one-dollar contract clears county budget scrutiny without a fight. It also buys a system nobody at the county has a financial reason to scrutinize, and that is where the honest problem shows up. More than seven months after Manatee’s drone went live, Government Technology reported that dispatchers had not deployed it once. Not a hardware failure. A dispatch-protocol failure, which is the harder one to fix, because it means the aircraft sat charged and ready while call-takers ran the scripts they already knew.

Hillsborough’s design partly answers that. Paramedics on scene request the delivery from their phones rather than dispatchers making the call cold, and Fire Rescue keeps operational control and the Certificate of Waiver authority. Whether that is enough to get the thing used is the question the pilot exists to answer.

Folkes Says the FAA Path Got Shorter Since 2024

Founder and CEO Gordon Folkes told a Sarasota-area newspaper in October 2025 that FAA changes had compressed the approval path for public safety flights. ArcherFRS operates at or below 200 feet (61 meters) with an ADS-B receiver aboard, a configuration he credits for shorter waiver waits.

“I’m grateful the FAA has created a clear path for public safety operations,” Folkes said.

Waiver waits that ran close to a year before those changes, by Folkes’s account, are what made the old version of this program impossible on a four-week clock. A program approved September 2 has exactly that long to stand up two hubs, and the regulatory piece being routine now is what makes that survivable. It also sits oddly against the Part 108 delay we covered in June, where the commercial beyond visual line of sight rule keeps slipping while public safety operators get moved through on waivers. Public aircraft status remains the fast lane.

Neither Tampa General nor ArcherFRS has stated what Hillsborough County would pay. The Manatee figure of one dollar a month is public because it went through a county contract, and the Hillsborough number will become public the same way when the agenda item posts. As of this writing it has not, which is unremarkable two weeks out but does mean every published account of this program so far, including ours, has described a deal whose terms nobody outside the room has read.

DroneXL’s Take

I want this program approved, and I think it will be. Drones carrying blood to people bleeding out on a Riverview roadside is the clearest possible answer to everyone who still thinks this industry is about flying cameras. Six years ago we wrote up ArcherFRS pitching Manatee County on AEDs and Narcan, waiting on an FAA that had not figured out what to do with them. Now they are moving blood products and the waiver that used to take a year does not. That is a decade of work landing.

So the criticism here is narrow and it is not about the drones. Announcing an unapproved program in the present tense, then watching trade outlets repeat it as fact for two weeks, is how the industry burns the credibility it needs at exactly these county-commission moments. Public safety drone programs live or die in rooms full of skeptical elected officials, and the fastest way to lose one is to have a commissioner discover that the vendor already told the national press the vote was a formality.

The honest counterweight, and it cuts against my own enthusiasm: Manatee’s drone sat unused for over seven months. Duke Health’s AED program and Everdrone’s documented cardiac save in Sweden prove the model works when the dispatch chain actually pulls the trigger. The aircraft was never the bottleneck. Human protocol was, and Hillsborough is buying a system whose hardest problem is not on the spec sheet.

My read is that the vote passes and October holds, because a hospital system absorbing the cost makes this close to free for the county and no commissioner wants to be the one who voted against blood. Watch September 2 for the number attached to it. If Hillsborough is paying something meaningfully more than Manatee’s dollar, the Champion Hospital model is being priced up as it scales, and every fire chief reading about this program should know that before they call Orlando.

Sources: Tampa General Hospital, Archer First Response Systems via PR Newswire, Government Technology, Your Observer, Firehouse / Sarasota Herald-Tribune

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

Neros Raises $250M At $2.5B As The Army Bets On A Million Drones

Haye Kesteloo — Tue, 18 Aug 2026 16:18:00 GMT

Neros Technologies closed a $250 million Series C on August 11 at a $2.5 billion post-money valuation, roughly triple the mark the company carried nine months ago. Sequoia Capital and the American Strategic Technology Fund co-led the round, joined by Interlagos, Valor Equity Partners, Allen & Company, Thiel Capital, Spark Capital and Figma co-founder Dylan Field.

When I profiled the two former teenage drone racers who founded the company last November, Neros had raised $121 million in total and was turning out roughly 2,000 drones a month. Nine months later it is valued at $2.5 billion and the line is producing about 1,200 aircraft a week.

The money funds two new products and a manufacturing target the founders committed to publicly back when the whole operation fit inside a 15,000-square-foot (1,394-square-meter) factory in El Segundo, California.

Sequoia Came Back For Its Second Consecutive Lead

Sequoia Capital led the $75 million Series B in November 2025 and returned to co-lead this one, alongside the American Strategic Technology Fund, an El Segundo firm that invests in US strategic technology. Interlagos, which also backed the Series B, joined the round again.

Neros went from a company that had raised $121 million across its entire life to a $2.5 billion company in a single round, nine months after the Series B rather than the eighteen to twenty-four months a raise of that size usually takes. CEO Soren Monroe-Anderson said the core business has grown fast enough since November to justify pulling it forward. “Our mission to produce one million drones per year hasn’t changed,” he said.

Washington has been circling the same cap table. In May, the Wall Street Journal reported that the Pentagon’s Office of Strategic Capital had identified Neros, Performance Drone Works and Unusual Machines as candidates for federal financing that could include equity, and DroneXL covered the structure of that proposal in full. Private capital got there first. For the background on how two 20-somethings reached this point, our November profile of the founders covers the Kyiv trip that set the company’s direction.

Archer AI And Bandit Push Neros Past One-Way Attack

Two products carry the new money. Archer AI keeps a human on the sticks but adds terminal guidance for the final few hundred meters and position hold for flying where GPS is jammed. Bandit is a counter-drone interceptor aimed at Class 2 and 3 threats.

That category includes Shahed-style aircraft, and Neros says both platforms will be in combat theaters by the end of 2026, according to the company’s announcement. Terminal guidance matters because the last stretch of an FPV attack run is where electronic warfare concentrates and where most drones are lost. Position hold covers the other half of that problem, letting an aircraft hold station after satellite navigation drops out.

Bandit is the harder bet of the two, and Neros is making it while the strike line is still climbing toward its own targets.

Both platforms run on hardware built for multi-asset control, the capability usually marketed as swarming. The company delivered its first units under the Army’s attritable-drone program ahead of schedule in March, launching the Archer Block 2 airframe with components swappable across five-, eight- and ten-inch configurations. An Archer runs about $2,000, or roughly $5,000 as a complete system with a warhead and support equipment.

The Army’s $500 Million Contract Sets A Ceiling, Not A Floor

The Army awarded Neros an indefinite-delivery contract on June 30 with a $500 million ceiling for Archer drones under the Purpose-Built Attritable Systems program, running through June 2031. A ceiling is not a commitment, and the service can order a fraction of that amount and stay within the terms.

Neros told Defense Daily the award could eventually cover hundreds of thousands of aircraft and that it has already begun filling a first order of thousands. The order is the part with an actual number attached, and that number is thousands, not hundreds of thousands. The Army is trying to lift annual small-drone purchases from around 50,000 units to more than a million, and the Pentagon’s Drone Dominance Program set out to buy more than 200,000 industry-built drones by 2027 when it launched in December 2025, a figure the War Department pushed past 300,000 a day later. Both goals flow from Secretary Pete Hegseth’s July 2025 directive and the memorandum itself.

Neither the Army nor Neros has stated a guaranteed minimum quantity under that contract, and the Series C announcement says nothing about how much of the million-a-year target is backed by orders rather than by capacity. The award sets work and funding per delivery order. That gap is the difference between a company with a million-drone factory and a company with a million-drone customer, and the public record currently establishes only the first.

Neros Has To Multiply Its Output Sixteen Times By 2028

About 1,200 Archers a week works out to roughly 62,000 a year. The 2028 target is one million. That is a sixteenfold increase inside roughly two years. The intermediate step Neros has named publicly is 10,000 drones a month by the end of 2026.

The company is building toward it in a 250,000-square-foot (23,226-square-meter) facility in Torrance, California, and expanding abroad through a UK subsidiary that put up to £10 million into a Swindon-area FPV hub in March. Sovereign production lines in allied countries are part of the stated plan.

When DroneXL wrote up DARPA’s heavy-lift prize last November, Neros was managing about 2,000 drones a month, a volume Chinese factories clear in two days. Neros has more than doubled its monthly rate since then and that comparison has barely moved. Every American production target is set against that gap, which is why the 2028 number reads as industrial policy rather than a forecast.

DroneXL’s Take

This is the company the pro-domestic-manufacturing argument was always supposed to produce, and its existence makes the case against the FCC drone ban stronger rather than weaker. A real American FPV manufacturer operating at scale removes the last excuse for banning the competition.

The strongest argument against that reading is that Neros is riding demand policy created. Executive Order 14307, the Hegseth memo, the tariffs I wrote about three days ago and the Covered List all pushed money toward exactly this kind of company. Nobody was building a $2.5 billion FPV manufacturer in 2019.

The position survives anyway. Monroe-Anderson and Hichwa flew to Kyiv to learn manufacturing from Ukrainians before the Pentagon would take a meeting, they built without Chinese parts before it was mandated, and what moved this valuation was units leaving a loading dock. Set that against the Powerus roll-up I took apart in March, where a 10,000-drones-a-month claim rested on a reverse merger and a shell company. Both stand to gain from the same policy. Only one of them is shipping.

My read is that this is exactly why banning DJI remains the wrong instrument. Neros did not need the FCC to hobble a Chinese competitor. It won Army and Marine Corps business by being the best NDAA-compliant FPV supplier available, in a segment DJI does not sell into at all. The ban does nothing for Neros. It takes capable, affordable hardware away from American public-safety agencies and surveyors who have no Blue UAS equivalent at anything near the price, and it lets a genuine domestic success story serve as cover for a policy that has nothing to do with it.

Watch the 10,000-a-month figure at the end of December. Neros published that number itself, which makes it the cleanest available test of whether the ramp is real. Hit it and the 2028 target stops looking like a slogan. Miss it badly and every million-drone projection in this industry, the Army’s included, deserves a harder look.

Sources: Neros Technologies, Defense Daily, GovConWire, DefenseScoop, US Department of War.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.

DroneXL: Drone News, DJI Rumors, Reviews, and Drone Laws

Oregon Warns Drone Pilots: One Flight Can Ground Every Fire Aircraft

Oregon’s Warning Follows Drone Delays at the Grasshopper Fire

Drone Incursions Have Grounded Aircraft Across the West This Month

Violators Face $75,000 Fines and Federal Prison Time

DroneXL’s Take

FCC Labels Drone Light Shows Military-Grade While US Operators Fly Korean And Chinese Fleets

The FCC Puts Light Shows Inside Its Swarm Category

American Show Operators Fly Korean-Built Aircraft

Chinese Manufacturers Supply Most Of The World’s Show Drones

Nova Sky Stories Manufactures In The US, Europe, And China

Creative Skies Tells Show Operators To File Before September 2

DroneXL’s Take

Saildrone Fires First Missiles From Its Robot Ship at US Navy's RIMPAC

Two Missiles Left the Rail Through the Navy’s ARES Combat System

Lockheed Martin’s October Investment Produced a Live Launch in Six Months

The 170-Foot Spectre Carries the Bigger Payload Ambition

DroneXL’s Take

DJI Mini 5 Pro Amazon Bundle Packs Over $100 In Free Gear For $899

The 1-Inch Sensor Is Still The Reason To Buy

Air Photography And Jake Sloan Reached The Same Verdict

Registration Applies No Matter Which Bundle You Buy

DroneXL’s Take

Amazon's Pool Drop Shows Why Zipline Is Winning Drone Delivery

Amazon Built 83 Pounds of Aircraft Around a 5-Pound Box

Zipline Split the Delivery Into Two Aircraft

Zipline’s Parachute Solved Rural Delivery First

DroneXL’s Take

The FCC's 'Military-Grade' Drone Definition Is Broad Enough to Catch Your DJI Neo 2

This Is a Public Notice, Not a Rulemaking — and That Matters

Seven Categories Define “Military-Grade”

A Capability Test Catches a Lot of Civilian Equipment

What the FCC Is Asking the Public

The Carve-Outs, and Where the FCC’s Own Logic Falls Apart

How to File a Comment

FCC Tells All Certification Labs To Tighten Drone Screening

The FCC Outsources Nearly All Equipment Authorization

Bad Grants Have Reached The Market Before

The Drone Rules Have Been A Moving Target Since December

The Guidance Itself Has Not Been Published

DroneXL’s Take

The DJI Osmo Pocket 4 Accessories I Actually Use: Protection, ND Filters, Mics, Batteries, and Car Rigs

Protect the Gimbal and Screen First

A Magnetic Arca-Swiss Base That Does Everything

ND Filters for the Highest Quality Footage

The Wide-Angle Lens for Walk-and-Talk

A Microphone Makes People Stick Around

Batteries: The Handle Works, but This Power Bank Is Better

The LUTs I Grade All My Pocket Footage With

Car Rigs: The Pocket 4 Is a Fantastic Rolling Camera

A Motorized Slider That Unlocks Something New

Insta360 X6 In-Depth Review: What's Actually New — and Where It Falls Short

A Shorter, Wider Body — and On-Screen Buttons I Don’t Love

New Replaceable Lenses — and the One I Scratched Anyway

Bigger Sensors: How the X6 Stacks Up Against GoPro and DJI

10-Bit, I-Log, and Dolby Vision

The New 4nm Chipset: 8K50, 5K Single-Lens, and InstaFrame 2.0

AI Director Now Runs on the Camera Itself

Low Light: Brighter, But Not Always Correct

Underwater With the New Dive Case: A Bit of a Situation

Internal Storage and a Completely Redesigned Battery

120-Megapixel Photos and AI Reframing

NFC Tap-to-Connect, Faster Bluetooth, Wi-Fi 6, and USB 3.0

Pricing — and the DJI Osmo 360 II Waiting in the Wings

Should You Upgrade From the X5?

DJI Offers $10,000 For Your Best Avata 360 Video While Washington Decides Its US Future

DJI Puts $25,000 In Prizes Behind The All For Passion Challenge

Entries Compete On Quality, Not Follower Count

The Avata 360 Reached US Buyers Late And Through Amazon

DroneXL’s Take

DJI Names 15 Onboard AI Challenge Winners, From Banana Farms to Bridge Cracks

Ten Judges Split 15 Winners Across Two Award Tiers

AgroCount AI Turns a Week of Banana Counting Into Onboard Math

Hangzhou New Modal Packs Nine Traffic Inspections Into One System

The Winner List Runs Straight Through China

DroneXL’s Take

DJI Scores Its Biggest Legal Win in Years, New Drone Tariffs Hit 100%, and the HOVERAir Versa Clears the FCC

DJI Scores Its First Big Legal Win in Years Against the Pentagon Blacklist

New Drone Tariffs as High as 100% Are Coming Within Weeks

The HOVERAir Versa Somehow Cleared the FCC While DJI Stays Blocked