OpenAI announced the rollout of GPT-5.5-Cyber in a limited preview alongside expanded Trusted Access for Cyber capabilities designed to support cybersecurity defenders securing critical infrastructure and enterprise systems.
The company said GPT-5.5-Cyber is intended for specialized cybersecurity workflows, including red teaming, vulnerability validation, penetration testing, malware analysis, binary reverse engineering, and detection engineering in controlled, authorized environments.
OpenAI stated that GPT-5.5 remains its primary recommended model for most cybersecurity workflows through Trusted Access for Cyber, while GPT-5.5-Cyber is designed to provide more permissive access for a smaller set of highly verified defenders and organizations.
Trusted Access for Cyber is an identity- and trust-based framework that reduces classifier-based refusals for verified cybersecurity defenders while continuing to block malicious activities such as credential theft, malware deployment, persistence, stealth operations, and exploitation of unauthorized third-party systems.
According to OpenAI, approved users and organizations will be required to implement phishing-resistant authentication protections beginning June 1, 2026.
The company said GPT-5.5 and GPT-5.5-Cyber are being deployed across multiple areas of the cybersecurity ecosystem, including vulnerability research, patch validation, software supply chain security, detection engineering, monitoring operations, and network defense.
OpenAI is also partnering with cybersecurity companies and infrastructure providers, including Cisco, Intel, SentinelOne, and Snyk, to evaluate how AI models can accelerate defensive workflows and large-scale protection systems.
In addition, OpenAI announced Codex Security, a new system designed to help open-source maintainers identify, validate, and remediate vulnerabilities through codebase-specific threat modeling, attack-path analysis, and patch recommendations.
The company expects future model releases to continue expanding cybersecurity capabilities and to incorporate stronger identity verification, approved-use scoping, monitoring systems, and proportional safeguards.
KEY QUOTES:
“At Cisco, we view frontier models as a powerful force multiplier for defenders. Models like GPT-5.5 are fundamentally changing the velocity of our operations, enabling us to move faster on everything from incident investigation to proactive exposure reduction. But speed cannot be traded for trust. The true value of this technology isn’t found in the model alone, but in the enterprise-ready framework we wrap around it.”
Anthony Grieco, SVP And Chief Security & Trust Officer, Cisco
“Intel is a leader in silicon and software, providing a trusted foundation for the global computing industry. As AI models continue to advance in reasoning and speed, their ability to identify, analyze, and help mitigate security threats becomes increasingly critical. Intel looks forward to partnering with OpenAI to bring governed, scalable AI capabilities into real-world cyber workflows.”
Dhinesh Manoharan, Head Of INT31 Security Research, Intel Corporation
“At SentinelOne, the real value of AI is how quickly it helps us turn signals into an actionable advantage for defenders. GPT-5.5 helps analysts connect telemetry, focus on what matters, and strengthen how organizations investigate, detect, and respond to emerging threats.”
Gregor Stewart, Chief AI Officer, SentinelOne
“Attackers are already weaponizing frontier models. By deploying OpenAI’s Trusted Access for Cyber and GPT-5.5, we are giving defenders at Snyk the capability they need to protect critical supply chains. This partnership isn’t just a milestone, it’s a strategic necessity.”
Manoj Nair, Chief Innovation Officer, Snyk
