Today, OpenAI announced the limited preview rollout of GPT-5.5-Cyber, aimed at defenders responsible for securing critical infrastructure. This model, positioned as an advanced tool for identifying and patching vulnerabilities, has been recognized by independent evaluations as one of the strongest for specialized cyber tasks. OpenAI emphasizes that GPT-5.5-Cyber is designed solely for defensive purposes, integrating safeguards to prevent misuse, and is accessible through its Trusted Access for Cyber program, which ensures that only vetted defenders can utilize its capabilities.

Fouad Matin: Fouad Matin is a product and policy lead at OpenAI focused on cybersecurity initiatives and the Trusted Access for Cyber (TAC) program. In this news, he is the one announcing the limited-preview rollout of GPT‑5.5‑Cyber to critical‑infrastructure defenders and positioning GPT‑5.5 with TAC as a primary option for developers to discover and patch vulnerabilities in their own code.
GPT-5.5-Cyber: GPT-5.5-Cyber is a specialized variant of OpenAI’s GPT-5.5 model, fine-tuned for cybersecurity workflows such as vulnerability identification, malware analysis, and reverse engineering under controlled conditions. In this news, it is being rolled out in a limited preview to vetted defenders protecting critical infrastructure, giving them a more permissive tool for authorized testing and defensive security operations within OpenAI’s Trusted Access for Cyber program.

`json
{
“Model_Capabilities”: “Independent evaluations describe GPT-5.5 as one of the strongest currently tested models on specialized cyber tasks, including complex multi-step attack and defense simulations.”,
“Defensive_Use_Focus”: “OpenAI analyses emphasize that even the more permissive GPT-5.5-Cyber variant is constrained to defensive and authorized testing uses, with safeguards that continue to block capabilities like credential theft and unauthorized malware deployment.”,
“Trusted_Access_Program”: “OpenAI’s Trusted Access for Cyber program gates its more cyber-capable models, including GPT-5.5-Cyber, behind identity verification and tiered access to ensure they are used by vetted defenders rather than attackers.”
}
`