Business News Tech AI Anthropic strengthens defences against distillation attacks against Claude, cuts data harvesting by upto 99%
Anthropic strengthens defences against distillation attacks against Claude, cuts data harvesting by upto 99%
Anthropic has introduced new security measures to prevent illegal model distillation attacks targeting its Claude models. The company reports a significant reduction in data harvested by unauthorized entities, with estimates ranging from 90 to 99 percent. These distillation attacks pose serious risks related to biosecurity, cybersecurity, and national security. Anthropic's approach includes restrictions on reasoning data and the use of advanced classifiers to block extraction attempts.
Artificial intelligence research company Anthropic has introduced advanced security measures to curb illicit model distillation attacks against its Claude models, reporting an estimated 90 to 99 per cent drop in data harvested by unauthorized entities. The developments, detailed in an Anthropic blog post, address growing concerns over large-scale industrial campaigns designed to siphon proprietary reasoning capabilities from frontier systems to train competitor models without authorization. "Distillation attacks are covert, industrial-scale campaigns to extract the capabilities of one model and replicate them in another, without the authorization of the original model's developers," Anthropic stated in its blog post.
"They are often conducted using networks of fraudulent accounts. These attacks allow companies to mimic frontier capabilities for a fraction of the development time and cost," the company noted. The report highlights that such practices pose severe biosecurity, cybersecurity, and national security risks. Because distilled variants often enter circulation without proportional safety guardrails, bad actors can exploit them for malicious purposes.
Source link
