Anthropic: 'We made the wrong tradeoff' in new model guardrails

Anthropic reversed its policy on Claude Fable 5’s secret safeguards after backlash from the AI community, making the rerouting of flagged requests to Opus 4.8 visible to users. The update is primarily a governance and product-transparency issue rather than a fundamental business or financial development. Market impact appears limited, though it may affect sentiment around Anthropic’s approach to AI safety and control.

Analysis

Anthropic reversed its policy on Claude Fable 5’s secret safeguards after backlash from the AI community, making the rerouting of flagged requests to Opus 4.8 visible to users. The update is primarily a governance and product-transparency issue rather than a fundamental business or financial development. Market impact appears limited, though it may affect sentiment around Anthropic’s approach to AI safety and control.