OpenAI Rolls Out GPT-5.5-Cyber to Compete With Anthropic’s Mythos
SAN FRANCISCO — OpenAI on Wednesday began a limited rollout of GPT-5.5-Cyber, a cybersecurity-focused AI model, to vetted security teams roughly a month after rival Anthropic launched its Mythos cyber model, according to CNBC.
The new model is being made available in limited preview to vetted cybersecurity teams, according to CNBC. The company is positioning GPT-5.5-Cyber as a direct competitor to Anthropic’s Mythos, a cybersecurity-focused AI model that launched approximately one month ago.
GPT-5.5-Cyber is a variation of OpenAI’s latest GPT-5.5 model, fine-tuned for cybersecurity use cases, according to CNBC. The release adds to competition among AI laboratories developing domain-specific models for security applications, with potential implications for U.S. government agencies, enterprise security operations and critical infrastructure protection.
The limited preview approach — restricting access to vetted teams rather than offering broad availability — reflects the sensitive nature of cybersecurity AI tools, which could potentially be misused if deployed without adequate safeguards.
Both OpenAI and Anthropic are U.S.-based AI laboratories. The rollout comes as both companies have expanded beyond general-purpose AI into specialized domain applications in recent months.
