A billion-dollar AI moat just took 28.8 million exchanges to crack, according to Anthropic. The company sent the claim straight to the US Senate, as investors question how long any AI lab’s edge can survive being copied this cheaply. Here’s what Anthropic alleges, and why it puts pressure on AI’s biggest price tags.Anthropic says this is bigger than a fight with one rival. Qwen allegedly used Claude’s answers to catch up fast, instead of doing the research itself. That means another country gets the benefit of America’s AI investment, without paying for any of it. A letter sent to senators on June 10 names Alibaba’s Qwen AI lab directly. It alleges operators tied to Qwen used nearly 25,000 fake accounts to run 28.8 million exchanges with Claude between April 22 and June 5, 2026. CNBC confirmed the letter exists. Anthropic calls it the largest case of this kind it has disclosed publicly, and says Alibaba acted “brazenly.” Can AI survive without copying ?
