Anthropic Releases Claude Fable 5 Model, Restricts Topics on Cybersecurity and Biology

According to Anthropic, the company publicly released Claude Fable 5 on Tuesday, a new Mythos-class model that surpasses its previous frontier Opus models in overall capabilities. The model comes with safeguards designed to prevent it from answering queries on sensitive topics including cybersecurity, biology, and chemistry, where Anthropic has expressed concerns about potential misuse by malicious actors. Unlike Mythos 5, which remains limited to a small group of cyberdefenders, Fable 5 is publicly accessible and routes restricted queries to the earlier Claude Opus 4.8 model while warning users of this redirection. Anthropic stated these safeguards were tuned to be stricter than ideal, resulting in less than five percent false positive rates in testing, which the company deems acceptable to mitigate risks of serious harm.
Disclaimer: The information on this page may come from third-party sources and is for reference only. It does not represent the views or opinions of Gate and does not constitute any financial, investment, or legal advice. Virtual asset trading involves high risk. Please do not rely solely on the information on this page when making decisions. For details, see the Disclaimer.
Comment
0/400
No comments