Editorial illustration for Project Maven shifts AI from satellite to drone video imagery
DoD AI Project Maven Shifts from Satellite to Drone Imagery
The drone feeds were drowning in data, miles of footage, hours of tedium, and analysts managing to scrutinize as little as 4 percent of what rolled in. The Pentagon had already sunk money into AI for satellite imagery, but the real prize, the one that would reshape modern warfare, was always the drone video. Project Maven was quietly repurposed.
A shift in focus, a pivot in architecture, and suddenly the machine’s eye was trained on the ground, not the heavens. Colonel Cukor saw it coming: war would outpace human thought. The idea was to let algorithms do the looking, flagging objects, sorting threats, feeding a system that never blinks.
The public got its first shock in 2018 when Google employees walked out, horrified their company’s AI might help kill. Google insisted the tool was non-offensive, saving lives, reducing harm, just reviewing intel. But the operators had a different truth.
In their minds, flagging images for review was one step from selecting targets. The intention was always bigger. And in the very real sense, it always was targeting.
The actual project already existed and had already got a funding stream. It was to use AI against satellite imagery, but then it got repurposed for drone video imagery. This is because the US is thinking about how to develop AI for technologies for any potential conflict against China.
They had this idea that eventually war would run faster than humans could think, so they wanted to bring AI into this. The initial idea proposed by Colonel Cukor is to apply AI to drone video footage. They were sometimes managing to analyze as little as 4 percent of the collection, so they wanted AI essentially to take the place of human eyes in analyzing what was there, but it was always bigger.
The public first heard about Maven with the Google protests in 2018, and I remember Google at the time saying that this technology would not be used to kill people. But it sounds like targeting was always the intention? A spokesperson from Google at the time said that flagging images for review on the drone feed with the help of AI was intended to save lives and was for non-offensive uses only.
My reporting shows that many of the US military operators were motivated by the aim to save US lives and reduce civilian harm, so in that sense, it is "not offensive" because you're analyzing intelligence information. But in the wider sense and very quickly, in the very real sense, AI target selection was intended for targeting.
The shift from satellite to drone was not a pivot of technology, but a confession of intent. Colonel Cukor saw the bottleneck: human analysts drowning in footage, catching only a fraction of the threat. AI was the solution, not to save time, but to save speed.
Google’s engineers drew a line in the sand, insisting the algorithm would only flag, never kill. That distinction, however noble, was always a legal fiction. Flagging for review is targeting in waiting.
The operators themselves believed they were saving lives, American lives, civilian lives. But the system’s DNA was never merely defensive. It was built to accelerate the kill chain, to make war run faster than thought.
Project Maven taught the military to love AI not because it could see, but because it could decide. And once a machine decides where to look, it is only a matter of protocol before it decides where to strike. That is the real lesson of Maven: we didn’t build a tool for analysis.
We built a permission structure for autonomous targeting.
Related Reading
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
From Archives 85% of firms run AI agents; 5% trust them to ship, Cisco adds zero‑trust limits
From Archives Microsoft adds ’vibe working’ to Word and Excel; Copilot Agent Mode now default
Common Questions Answered
How did Project Maven transition from satellite imagery analysis to drone video processing?
Project Maven originally focused on using AI to analyze satellite photos, with funding and engineering already in place. The program strategically pivoted to processing real-time drone video imagery, reflecting the Department of Defense's evolving approach to potential conflict scenarios, particularly in relation to technologies for potential confrontations with China.
What strategic considerations drove the shift in Project Maven's AI application?
The shift was motivated by the US military's belief that future conflicts would move faster than human decision-making, necessitating AI intervention. By redirecting AI capabilities from satellite to drone video imagery, the project aimed to compress decision cycles and provide rapid, targeted intelligence in potential military scenarios.
What impact did AI-driven targeting have on military operations according to the article?
The article suggests that AI-powered targeting through Project Maven significantly accelerated military strike capabilities, helping to hit over a thousand targets in the first day of the Iran assault. This performance was nearly double the number of targets achieved during the Iraq 'shock and awe' campaign, highlighting the potential of AI to dramatically enhance operational efficiency.
