OpenAI says it is widening its cyber defense program by naming the first organizations receiving support from its Cybersecurity Grant Program and by expanding access to GPT-5.4-Cyber through its Trusted Access for Cyber program.

What OpenAI disclosed

In a new ecosystem update, OpenAI said it has committed $10 million in API credits for defensive security work. The first named recipients are Socket, Semgrep, Calif, and Trail of Bits, groups focused on software supply chain security, vulnerability research, and open-source defense.

OpenAI also said its Trusted Access for Cyber program is scaling to thousands of verified individual defenders and hundreds of teams responsible for protecting critical software. That rollout builds on this week's introduction of GPT-5.4-Cyber, a version of GPT-5.4 tuned for defensive cybersecurity tasks and offered on a limited, verified-access basis.

Why it matters

The update turns an earlier model launch into a broader ecosystem push. OpenAI said organizations including Cloudflare, CrowdStrike, NVIDIA, Oracle, Palo Alto Networks, and Zscaler have signed on to support the effort, while the U.S. Center for AI Standards and Innovation and the UK AI Security Institute have received access for evaluations focused on cyber capabilities and safeguards.

That makes the announcement more than a model release. OpenAI is tying frontier cyber models to grants, partner validation, and external testing, a sign that access to higher-risk defensive AI tools may increasingly be distributed through trust and verification programs rather than open self-serve release.