AIThe Decoder1h ago

Anthropic says Fable 5 has invisible safeguards

Anthropic says Fable 5 has invisible safeguards that use prompt modification, steering vectors, or PEFT to limit its effectiveness for building frontier LLMs

Anthropic says Fable 5 has invisible safeguards

Key Points … Ask about this article... Both models share the same base model. Fable 5 ships with conservative safety guardrails for general use.

Read full article

Source: The Decoder · Opens in new tab