Accelerating a Cyber Defense Ecosystem That Protects Us All
Key point
OpenAI is expanding Trusted Access for Cyber to provide defense teams and institutions with advanced cyber capabilities.
Details
Trusted Access for Cyber is a concept aimed at broadly providing advanced cyber capabilities to more defenders, while expanding access in stages based on trust, verification, and safeguards. OpenAI unveiled the first organizations to actually apply this approach, emphasizing broad participation ranging from open-source security teams and vulnerability researchers to enterprises operating complex digital environments.
Because cybersecurity is an area where many stakeholders move together, this program targets an ecosystem that encompasses not only large enterprises and security vendors but also researchers, maintainers, public institutions, nonprofits, and small teams. OpenAI stated that, because every software developer should be able to leverage the defensive capabilities of frontier models, it is committing $10 million in API credits to the Cybersecurity Grant Program.
Initial beneficiary organizations include Socket and Semgrep, which focus on software supply chain security, and Calif and Trail of Bits, which combine frontier models with vulnerability research expertise. OpenAI explained that it plans to further collaborate with teams that have a proven track record of finding and fixing vulnerabilities in open-source software and critical infrastructure systems, and that it is also accepting additional applications.
Major organizations in finance, security, and infrastructure are already participating in the program. The disclosed participating companies and institutions include Bank of America, BlackRock, BNY, Citi, Cisco, CrowdStrike, Goldman Sachs, iVerify, JPMorgan Chase, Morgan Stanley, NVIDIA, Oracle, SpecterOps, and Zscaler.
OpenAI has also provided GPT-5.4-Cyber to the U.S. Center for AI Standards and Innovation (CAISI) and the UK AI Security Institute (UK AISI) to evaluate the model's cyber capabilities and safeguards. OpenAI stated that it will improve its safety framework based on lessons learned from real-world use, and will continue to expand Trusted Access for Cyber so that defensive capabilities can be put to good use across the broader ecosystem.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.