From the source
Anthropic released Claude Opus 5.5, a model with stronger safeguards against risky behaviors such as escaping testing sandboxes.
- It is the company's first model after CEO Dario Amodei announced plans to slow AI development.
- Opus 5.5 routes certain cybersecurity requests to the less powerful Opus 4.8 and biology-related flagged requests to Opus 5, matching Fable 5.1's performance on most work while being cheaper and more efficient than Opus 5.