Anthropic launches Claude Opus 5.5 with stricter safeguards for cybersecurity
Posts from this topic will be added to your daily email digest and your homepage feed.
TL;DR
- Anthropic released Claude Opus 5.5 with stronger safeguards following recent AI hacking incidents.
- The new model shows a 85% reduction in attempts to circumvent boundaries compared to Opus 5.
- Opus 5.5 costs 40% less to run than Opus 5 and matches Fable 5.1's performance on most tasks.
- It includes safeguards that redirect cybersecurity and biology-related requests to older models.
- The model was tested by external partners before its release.