The broad safeguards built into Anthropic's Claude Fable 5, a Mythos-class AI model, blocks some mundane requests on cybersecurity and biology.
Why Anthropic's 'safe' Mythos-class model won't answer questions about cancer
By
Kelsey Vlamis
You're currently following this author!
Want to unfollow? Unsubscribe via the link in your email.
New frontier model refuses cybersecurity, biology, and chemistry queries.
Anthropic has unveiled Claude Fable 5, a model similar to Claude Mythos but with major cyber barriers blocking risky prompts and…
Anthropic has released a full version of its cybersecurity-centric Claude Mythos model—along with a safer version for the general…
Anthropic launches Claude Fable 5, a Mythos-class AI model engineered with new safeguards that specifically restrict its use in…
Anthropic's Claude Fable 5 routes basic biology and cybersecurity questions to an older model by design, raising concerns for…
Anthropic launched Claude Fable 5 with safeguards for high risk tasks while limiting Mythos 5 to trusted cyber partners.
