DJI

Based on Wikipedia: DJI

In the skies over Ukraine, cheap consumer drones are rewriting the rules of modern warfare. The company that makes most of them—over ninety percent of the world's consumer drones, in fact—was started by a college student building gadgets in his Hong Kong dormitory. That company is DJI, and its story reveals something profound about how power shifts in the twenty-first century.

From Dorm Room to Global Dominance

Frank Wang was studying at the Hong Kong University of Science and Technology in 2003 when he started tinkering with flight control systems. He wasn't trying to revolutionize warfare or surveillance. He just wanted to build things that could fly.

His first customers were universities and Chinese electric companies who bought the flight control components he assembled in his cramped student quarters. The money was modest, but it was enough to relocate to Shenzhen—the sprawling manufacturing metropolis just across the border from Hong Kong that had become the hardware capital of the world.

The early years were brutal. Wang had what colleagues diplomatically described as an "abrasive personality" combined with perfectionist expectations. Employees churned through the company at alarming rates. The operation survived largely because of Lu Di, a friend of Wang's family, who invested ninety thousand dollars and handled the finances while the young founder focused obsessively on engineering.

A turning point came in 2009. A team using DJI's flight control components successfully piloted a drone around the peak of Mount Everest. It was a proof of concept that these systems could work in the most demanding conditions imaginable—thin air, brutal cold, unpredictable winds. If a drone could handle Everest, it could handle just about anything.

The Phantom Revolution

Before 2013, drones were the province of hobbyists and professionals. Flying one required technical knowledge, patience, and a willingness to crash expensive equipment repeatedly while learning. The learning curve was steep, the costs were high, and the potential audience was limited.

Then came the Phantom.

Priced at six hundred twenty-nine dollars, the DJI Phantom was the first drone that regular people could actually use. It came fully assembled—no soldering, no configuration, no weekend-long build sessions. You charged the battery, downloaded the app, and flew. The software handled the complexity. The user just pointed where they wanted to go.

This was the iPhone moment for drones. Just as Apple had taken the smartphone from corporate executives and tech enthusiasts and put it in everyone's pocket, DJI took the drone from model aircraft clubs and military contractors and put it in the hands of real estate agents, wedding photographers, and curious teenagers.

The impact was immediate and devastating—to DJI's competitors, that is. By 2015, the Phantom 3 had added live-streaming video, letting pilots see exactly what their drone's camera saw in real time. Competitors simply couldn't keep up. Many went out of business. Others retreated to niche markets. DJI became, essentially, the only game in town.

The Hollywood Connection

The entertainment industry was quick to recognize what DJI had created. Those sweeping aerial shots that once required helicopters and specialized camera crews? Now a single operator with a Phantom could capture them. The economics were irresistible.

By 2017, DJI's technology had become so integral to television production that the company won a Technology and Engineering Emmy Award. The list of shows using DJI drones reads like a streaming service's homepage: The Amazing Race, American Ninja Warrior, Better Call Saul, Game of Thrones. Those soaring shots of Westeros? Largely captured by DJI equipment.

That same year, Frank Wang became Asia's youngest tech billionaire. He was also, more curiously, the world's first drone billionaire—a category of wealth that hadn't existed a decade earlier.

The Shenzhen Advantage

To understand DJI's dominance, you have to understand Shenzhen.

The city sits at the heart of the Pearl River Delta, the most concentrated manufacturing ecosystem on Earth. Within a few hours' drive, you can find factories making virtually every electronic component imaginable. Need a new motor design? There's a factory for that. Custom battery cells? Down the street. Specialized carbon fiber? Across town.

This density creates a feedback loop that's almost impossible to replicate elsewhere. DJI can prototype a new component, test it, revise it, and put it into production in a timeframe that would seem absurd to manufacturers elsewhere in the world. The company's Shenzhen factories feature sophisticated automated assembly lines, many built with equipment they designed themselves.

The workforce helps too. DJI employs roughly fourteen thousand people and is notorious for an internal culture that pits teams against each other in product development competitions. It's grueling. It's probably not pleasant. But it produces results.

When Toys Become Weapons

Here's where the story takes a darker turn.

A consumer drone is, at its core, a flying platform with a camera and a stabilization system. It can carry a small payload. It can hover in place. It can be operated from a distance by someone watching a live video feed. These are exactly the capabilities that make drones useful for aerial photography.

They're also exactly the capabilities that make drones useful for warfare.

The realization came gradually, then all at once. Police departments discovered they could use DJI drones for surveillance and search-and-rescue operations. By 2020, about ninety percent of drones used by American public safety agencies came from DJI. Fire departments used them to survey burning buildings before sending in crews. Search teams used them to cover ground that would take humans hours to traverse on foot.

But drones also found their way into less sanctioned hands. Terrorist groups began experimenting with consumer drones for reconnaissance and, occasionally, as improvised delivery systems for explosives. DJI implemented geofencing—software that prevents drones from flying in restricted areas—and other measures to limit misuse. But the fundamental problem remained: the same features that made a drone useful for filming a wedding made it useful for other purposes entirely.

The Xinjiang Question

In 2017, DJI signed a strategic cooperation agreement with Chinese police in Xinjiang.

That sentence requires context. Xinjiang is the western region of China home to the Uyghur people, a predominantly Muslim ethnic minority. The Chinese government has conducted what it describes as a counter-terrorism campaign in the region. Human rights organizations and Western governments have described it as something much worse: a systematic campaign of mass detention, surveillance, and cultural erasure affecting more than a million people.

DJI's agreement meant its surveillance drones would be used as part of this apparatus. The company has not been particularly forthcoming about the details.

This is the uncomfortable reality of DJI's position. It's a Chinese company, subject to Chinese law, operating in a political system where certain kinds of cooperation aren't optional. The company denies being state-owned, pointing out that state-affiliated investors hold less than six percent of ownership and less than one percent of voting rights. But ownership percentages don't fully capture the relationship between large Chinese tech companies and the Chinese state.

Ukraine: The Drone War

The Russian invasion of Ukraine in February 2022 turned DJI drones into weapons of war on an industrial scale.

Both sides use them. Ukrainian forces employ DJI drones for reconnaissance, artillery spotting, and propaganda—capturing footage of destroyed Russian columns that circulates on social media within hours. Russian forces do the same. The front lines are saturated with these devices, each one a pair of eyes in the sky for whichever side controls it.

This creates an almost absurd situation. A company that started making toys for aerial photographers now has its products being used to guide artillery fire and identify targets for destruction. The Phantom and Mavic drones designed for wedding videographers are helping decide the fates of soldiers.

DJI has attempted a kind of neutrality, announcing it would stop selling to both Russia and Ukraine. But the secondary market doesn't care about corporate pronouncements. Drones flow to both sides through various channels, and DJI's software updates—which might theoretically be used to disable units—seem to have limited practical effect in a war zone.

American Anxiety

The United States government has grown increasingly nervous about DJI.

The concerns center on data. Every DJI drone sends information back to company servers: flight paths, locations, video footage. The company says this data is used to improve products and is stored securely. Critics worry about what else might be done with it—especially given Chinese laws that can compel companies to share data with intelligence services.

In 2020, the Department of the Interior grounded around eight hundred DJI drones it had been using for wildlife conservation and infrastructure monitoring. The concern wasn't that the drones didn't work. They worked beautifully. The concern was what happened to the information they collected.

The U.S. government has since designated DJI as a "Chinese Military Company"—a label that carries various legal restrictions. The company has been sanctioned. American investment in DJI is restricted.

But here's the catch: you can still buy DJI drones in America. You can still fly them. The restrictions target government use and investment flows, not consumer purchases. As of 2020, DJI still held seventy-seven percent of the American consumer drone market. No competitor held more than four percent.

The gap between American security concerns and American consumer behavior tells you something about how difficult it is to disentangle from a dominant technology provider, especially when that provider offers a demonstrably superior product at a competitive price.

The Product Empire

DJI's dominance extends beyond the flying drones most people know.

The company makes gimbals—motorized stabilization systems that keep cameras steady during movement. The Ronin series has become standard equipment in professional video production. Watch almost any smoothly gliding shot in a modern film or television show, and there's a good chance a DJI gimbal was involved.

They make handheld cameras. The Osmo Pocket is a tiny, pocket-sized device that combines a camera with a three-axis gimbal, producing stabilized 4K video from something you can hold in one hand. Action cameras compete directly with GoPro. Smartphone gimbals turn shaky phone footage into smooth, professional-looking video.

They make agricultural drones. The Agras series can spray crops autonomously, carrying payloads of up to eighty kilograms. These aren't toys—they're industrial equipment, used by farms around the world to apply pesticides and fertilizers more precisely than human-operated equipment can manage.

They even, as of 2025, make robot vacuum cleaners. The DJI Romo represents an expansion into consumer robotics beyond flying machines. It's a reminder that the company's core competency isn't drones specifically—it's electromechanical systems, computer vision, and autonomous navigation. Those skills transfer.

The Camera Partnership

In 2015, DJI announced a strategic partnership with Hasselblad, the Swedish camera manufacturer whose medium-format cameras are legendary among photographers. Hasselblad cameras went to the moon with Apollo astronauts. They're the preferred tool of many of the world's best portrait and landscape photographers.

By 2019, DJI had acquired a majority stake in the company.

The partnership shows up directly in DJI's products. The Mavic 3 Pro, released in 2023, features a Hasselblad camera sensor. The newest Mavic 4 Pro, announced in May 2025, carries a hundred-megapixel Hasselblad sensor—resolution that would have seemed absurd for a consumer drone just a few years ago.

This acquisition pattern—identifying complementary technology and bringing it in-house—mirrors the strategies of other tech giants. Apple acquires camera technology companies to improve the iPhone. Google acquires artificial intelligence labs. DJI acquires camera expertise to make better flying cameras.

The Fraud Scandal

In January 2019, DJI announced something unusual for a Chinese tech company: it had discovered extensive internal fraud.

Employees had been inflating the costs of parts and materials, pocketing the difference. The scheme was sophisticated enough that it took an internal investigation to uncover. The estimated cost: up to one billion Chinese yuan, roughly one hundred forty-seven million U.S. dollars.

The disclosure was notable less for the fraud itself—large companies regularly discover internal corruption—than for the public acknowledgment. Chinese companies typically handle such matters quietly. DJI's willingness to announce the problem publicly suggested either a commitment to transparency or an attempt to reassure investors ahead of a potential public offering.

That offering, incidentally, has been rumored for years. In 2018, the company raised roughly one billion dollars in preparation for an initial public offering on the Hong Kong Stock Exchange. As of mid-2020, rumors persisted, but no IPO had materialized. The company's complex relationship with the American market—simultaneously dominant and sanctioned—may complicate any public listing.

The Workforce Factory

DJI's hiring process is legendarily difficult. The company recruits heavily from top Chinese universities and puts candidates through rigorous technical assessments. Those who make it through find an internal culture designed to be competitive to the point of discomfort.

Teams working on the same product category are sometimes deliberately pitted against each other. Multiple groups might develop competing designs for the same component, with only the winner going into production. It's an approach that maximizes innovation at the cost of collaboration—and probably employee wellbeing.

This cultural intensity helps explain both DJI's technical excellence and its employee churn problems. The company produces remarkable products. It also burns through people. Whether this trade-off is sustainable in the long term remains to be seen.

The Educational Turn

In 2019, DJI released something unexpected: a robot designed to teach kids to code.

The RoboMaster S1 is a tank-like ground rover that users must assemble themselves from loose parts. Once built, it's controlled through a first-person view interface via smartphone or computer. The educational angle comes from programming—users can code the robot's behavior using either Scratch, a visual programming language designed for children, or Python, the professional language used by software developers worldwide.

The product connects to DJI's RoboMaster competition, an annual international collegiate robot combat tournament the company has sponsored since 2015. Held at the Shenzhen Bay Sports Centre, it brings together university teams to build and fight robots. The S1 serves as both a training tool for aspiring competitors and an unofficial mascot for the event.

This educational investment might seem tangential to DJI's core business. But it makes strategic sense. Today's kids learning to code with a DJI robot might become tomorrow's engineers—or tomorrow's customers. The pipeline runs in both directions.

The Regulatory Maze

DJI's legal troubles extend beyond American security concerns.

In 2022, Textron—the American conglomerate that owns Bell helicopters and Cessna aircraft—sued DJI for patent infringement. The claim centered on automatic hovering technology: the systems that let a drone hold its position without constant pilot input. In April 2023, a U.S. jury agreed with Textron, ordering DJI to pay two hundred seventy-nine million dollars in damages.

DJI challenged the underlying patent, arguing it was obvious given prior art—meaning the technology was already known before Textron patented it. The Patent Trial and Appeal Board denied the challenge.

This kind of patent litigation is common in the tech industry. What makes the DJI case notable is the size of the judgment and the strategic implications. If competitors can successfully claim patent violations on core drone technologies, DJI's dominance could face legal as well as commercial challenges.

The Pandemic Pivot

When the coronavirus pandemic struck in 2020, DJI found unexpected uses for its technology.

In China, police used DJI drones equipped with speakers to remind people to wear masks. The image was dystopian and oddly comical: flying robots scolding pedestrians from above for failing to cover their faces. But it illustrated the surveillance capabilities these devices offered.

Other countries used drones for more technical purposes. In Morocco and Saudi Arabia, drones equipped with thermal cameras monitored human temperatures in public spaces, attempting to identify people with fevers who might be infectious. Others sprayed disinfectant over urban areas, though the effectiveness of aerial disinfection remained debatable.

The pandemic demonstrated something important about DJI's technology: its flexibility. A device designed for aerial photography could become a public health tool with minimal modification. The platform was more important than any single application.

The Whip Hand

The title of the Substack article that prompted this exploration—"China Has The Whip-Hand In The War"—refers to a fundamental truth about modern conflict. Wars are won through production and logistics. They're lost through lack of access to production and logistics.

DJI represents a specific case of this general principle. When ninety percent of the world's consumer drones come from a single company, in a single country, that country has leverage. Not just in military terms—though the Ukraine conflict makes that dimension impossible to ignore—but in commercial, technological, and strategic terms as well.

The United States has tried to develop alternatives. Various American drone companies have emerged, some with explicit government backing. None has come close to matching DJI's combination of capability, price, and ecosystem. The gap isn't closing.

This is what technological dependence looks like in practice. It's not abstract. It's a specific company, in a specific city, making specific products that the rest of the world has come to rely on. The question is what to do about it—and whether doing something is even possible at this point.

What Comes Next

DJI continues to push forward.

The Mavic 4 Pro, released in May 2025, represents the current state of the art: a hundred-megapixel Hasselblad sensor, a gimbal that can rotate three hundred sixty degrees (DJI calls it "Infinity"), fifty-one minutes of flight time, and obstacle avoidance in every direction. It's a remarkable machine.

The company's expansion into robot vacuums suggests ambitions beyond flying. The educational robotics line hints at a long-term vision for training the next generation of engineers. The camera technology from Hasselblad points toward ever more sophisticated imaging capabilities.

But the geopolitical shadow lengthens. American restrictions may tighten further. European regulators are watching. The use of DJI drones in surveillance and warfare creates liabilities that peaceful drone photography never anticipated.

Frank Wang started building flying machines in a dormitory. Two decades later, his company's products help determine the outcomes of wars. That arc—from hobby project to geopolitical flashpoint—says something about how quickly power can shift in a world where technology moves faster than institutions can adapt.

The drones keep flying. The questions about what they're watching, and for whom, remain unanswered.