OpenAI releases GPT-5.5-Cyber to rival Anthropic's Mythos
What's the story
OpenAI has announced the limited preview of GPT-5.5-Cyber, a specialized version of its latest artificial intelligence model. The move comes just a month after rival company Anthropic's successful launch of Claude Mythos Preview. The new model is being rolled out to selected cybersecurity teams and is designed to be more permissive on security-related tasks, as per an official blog post by OpenAI.
Enhanced capabilities
Model designed to assist vetted teams with advanced workflows
The new model, GPT-5.5-Cyber, is not a major upgrade in terms of cyber capability but is designed to assist vetted teams with advanced workflows where specialized access behavior may be important. These include tasks like vulnerability identification and triage, patch validation, and malware analysis. The company said that the general safeguards of the publicly available GPT-5.5 model would have made these tasks more difficult without this specialized version.
Competitive landscape
Anthropic launched cybersecurity-focused model last month
Anthropic, another major player in the AI space, had launched its own cybersecurity-focused model called Claude Mythos Preview last month. The company limited access to a select group of companies as part of a new initiative called Project Glasswing. This move drew attention from high-profile figures such as Federal Reserve Chairman Jerome Powell and Treasury Secretary Scott Bessent, who discussed Mythos with top US bank CEOs last month.