AIAnthropic1h ago
Anthropic says new measures like preserved thinking in Fable 5.1
Anthropic says new measures like preserved thinking in Fable 5.1 and Opus 5.5 have reduced data harvesting through distillation attacks by about 90% to 99%
TL;DRAnthropic reduced distillation attack success rates by 90-99% using new protective methods.
Why it matters: Harder to steal proprietary model capabilities means safer AI systems and protected competitive advantage.
In recent months, we've made substantial progress on reducing the effectiveness of distillation attacks on Claude using new methods, as we outline in this blog.
Read full articleSource: Anthropic · Opens in new tab