Asia

Anthropic accuses Alibaba of 'largest known distillation attack' on Claude AI model

US AI company Anthropic has accused Chinese rival Alibaba of unlawfully extracting capabilities from its Claude model. The firm described it as the 'largest known distillation attack'. Alibaba has not yet issued a detailed response.

A server room in a data centre
A server room in a data centrePhoto: panumas nikhomkhai / Pexels
Nikkei Asia1 h agoBABA

US-based artificial-intelligence company Anthropic has accused its Chinese rival Alibaba of extracting capabilities from its Claude model without authorisation.

Anthropic described the incident as the 'largest known distillation attack'. Distillation refers to using one model's outputs heavily to train another model, a practice that can breach terms of service.

The allegations come as competition between the US and China in artificial intelligence intensifies. Alibaba has not issued a detailed statement, and industry figures said disputes over intellectual property and model security are likely to grow.

AITechRegulationBABAAsiaNikkei Asia
This article is an AI-curated summary of the original story published by Nikkei Asia. The illustration is a stock photo by panumas nikhomkhai from Pexels and is not from the original story.

Read next