# Palantir's CTO Warns China Is Stealing AI, but the Real Threat Is at Home > Source: > Published: 2026-07-16 04:55:18+00:00 *Palantir's chief technology officer says China built its best AI models by siphoning answers out of American ones. He's less worried about that theft than about Americans turning against the data centers built to stay ahead of it.* Shyam Sankar sat down with Bloomberg Television on Wednesday. He said the quiet part out loud. "These Chinese open source models are really the result of distillation attacks," he told the network, describing them as built on "stolen American IP" from frontier labs like OpenAI, Anthropic and Google. It's a blunt accusation from a company whose business is selling AI-powered software to the Pentagon and its allies. Distillation itself isn't exotic. It's how a developer trains a smaller, cheaper model on the outputs of a bigger one, letting it inherit much of the larger model's behavior without the underlying training cost. Anthropic made the sharpest version of this argument two weeks earlier, telling the White House that operators tied to Alibaba's Qwen lab opened roughly 25,000 fake accounts and ran 28.8 million exchanges through Claude between April and June, an effort Anthropic labeled "adversarial distillation" aimed squarely at Claude's strongest skills: agentic reasoning and long, multi-step coding tasks. Alibaba has not addressed the specifics. DeepSeek and MiniMax have faced similar allegations. None of it has been independently verified. ## The Real Risk Isn't Copying, Sankar Says Here's the turn in Sankar's argument, and it's the more interesting one. He said the bigger economic risk to the US isn't Chinese copying at all. It's Americans refusing to build the data centers that keep frontier labs ahead in the first place. He compared the current backlash to the one that gutted US nuclear power in the 1970s, when Three Mile Island and a wave of local activism turned a technology with broad public support into a stalled industry for a generation. The comparison isn't hollow. Data Center Watch tracked 75 projects worth $130 billion in planned US construction blocked or delayed in just the first three months of 2026, and Gallup found 71% of Americans oppose a data center in their own area, more than the 53% who object to a nuclear plant nearby. Sankar's point: distillation is a copying problem. A stalled data center is a capacity problem, and capacity is the thing that actually decides who wins. ## A Convenient Argument for Palantir He's not a neutral narrator here. Palantir's stock and its government contracts depend on the US staying ahead in AI and keeping the compute pipeline flowing, so warning Washington against dampening data center growth doubles as an argument for Palantir's own future. That doesn't make him wrong. It does mean his testimony should be read as advocacy, not just analysis. Frankly, calling distillation "theft" is also more contested than Sankar lets on. Training on another model's outputs sits in a legal and ethical gray zone that predates this fight by years, closer to standard model-building practice than to industrial espionage, and American labs have used similar techniques against each other's outputs before. Anthropic's own complaint to the White House carefully avoided the word theft, opting instead for violations of its usage terms. What's harder to dispute is the price gap. Chinese labs have repeatedly shipped models that rival GPT-4 or Claude-class systems at a fraction of the training cost, a claim DeepSeek made about its R1 model earlier this year that rattled Nvidia's stock and reset assumptions about how much compute frontier AI actually requires. Whether that efficiency came from genuine research breakthroughs or from harvesting American outputs is exactly the question Sankar and Anthropic are now pressing Washington to answer. None of it resolves this year. But the direction is clear enough. Expect more of these public accusations timed to the same argument Sankar just made on camera: that keeping the US ahead means building more, not policing borrowed code. **Also read:** [TSMC posts record revenue and hands Wall Street its verdict on the AI buildout](https://startupfortune.com/tsmc-posts-record-revenue-and-hands-wall-street-its-verdict-on-the-ai-buildout/) • [What Is an AI Browser and How Do Agentic Tools Like Comet Work](https://startupfortune.com/what-is-an-ai-browser-and-how-do-agentic-tools-like-comet-work/) • [Bora Pharmaceuticals and Insilico Medicine Strike a $2.5 Billion AI Alliance](https://startupfortune.com/bora-pharmaceuticals-and-insilico-medicine-strike-a-25-billion-ai-alliance/)