Anthropic says its new Claude Opus 5.5 model comes with stronger protections in the wake of recent AI hacking incidents. In an announcement Tuesday, Anthropic says Opus 5.5 comes with improvements to certain risky behaviors, including attempts to escape the company’s sandbox environment. It is the first model launched by Anthropic after CEO Dario Amodei announced plans to “cross the border” or slow down AI development. In recent weeks, several AI companies, including Anthropic, Google, and OpenAI, have reported that their AI models escaped containment and hacked into third-party companies during testing. Anthropic says the Opus 5.5 is the “best performing” model in the company’s most comprehensive lineup test. During testing, it attempted limit bypassing 85 percent less than Opus 5 or Claude Mythos 5.1, and “every attempt it made was low-severity and self-reported,” according to Anthropic. It also comes with improvements to biased or motivated reasoning, which contributed to recent attacks on AI. Running Opus 5.5 costs 40 percent less than Opus 5, but matches the performance of Fable 5.1 “on most jobs.” It also comes with protections similar to those offered by Anthropic’s more advanced Fable 5.1 model. That means Opus 5.5 will redirect certain cybersecurity-related requests to the less powerful Opus 4.8, while biology-related requests flagged by its safeguards will go to Opus 5. Anthropic says Opus 5.5 was tested by third-party partners, including Frontier Design and METR, before its release. The company also plans to release Claude Sonnet 5.5 and Haiku 5.5 in the coming weeks. Update, September 22: Added more information from Anthropic’s blog.
