tech
Anthropic accuses Alibaba of campaign to 'brazenly' and 'illicitly' extract AI capabilities
The letter, which was obtained by CNBC, claims Alibaba carried out "the largest known distillation attack on Anthropic to date."
TL;DR
- Anthropic sent a letter to the U.S. Senate Committee on Banking, Housing, and Urban Affairs accusing Alibaba of illicitly attempting to extract its AI capabilities.
- The letter states Alibaba carried out "the largest known distillation attack on Anthropic to date," involving 28.8 million exchanges from roughly 25,000 fraudulent accounts.
- Distillation is an AI training method where a smaller model is built using outputs from a stronger model.
- Anthropic is working with Congress and the Administration to combat illicit distillation and maintain American AI leadership.
- This occurs two months after the White House pledged to help AI companies detect and coordinate against industrial-scale distillation.
- Anthropic previously identified similar, industrial-scale distillation campaigns from DeepSeek, Moonshot, and MiniMax.
- Separately, Anthropic received an export control directive from the Trump administration to suspend foreign nationals' access to its latest Claude models, citing national security authorities.