SecurityReutersabout 3 hours ago
Anthropic Cuts AI Distillation Attacks by 90%
2 min read
Anthropic reports new safeguards reduce AI distillation attacks by 90% to 99%. The company blocked tens of thousands of accounts, including Chinese startup Z.ai, which tried to extract Claude capabilities for its GLM-5.3 model. Reasoning token extraction attempts dropped 10 to 100 times compared to previous levels.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
