OpenAI and 100+ Companies Back Global AI Cyber Defense Plan

OpenAI is calling for a coordinated global cyber defense push as increasingly capable AI models make cyberattacks faster, cheaper, and more sophisticated.

The company published an open letter signed by more than 100 organizations, including Microsoft, Google, Anthropic, AWS, AMD, Cisco, Cloudflare, CrowdStrike, IBM, Oracle, Palo Alto Networks, and Fortinet.

Critical infrastructure faces growing AI security risks

The signatories warn that hospitals, water treatment facilities, internet infrastructure, and other essential services could face particularly serious risks.

Many organizations still depend on legacy systems, unpatched software, weak authentication, and accounts with excessive permissions. More capable AI models could help attackers discover these weaknesses and exploit them across large numbers of systems.

The companies want organizations to identify their most dangerous vulnerabilities first and strengthen access controls before AI-powered attacks become more common.

OpenAI wants defenders to get access to powerful AI

The letter also argues that cybersecurity professionals need access to capable AI models.

Security teams could use AI to find vulnerabilities, develop patches, analyze attacks, and respond to incidents more quickly. Cybersecurity vendors should also test their products continuously against frontier AI capabilities and add AI-powered defenses to existing security platforms.

The goal is to give defenders the same advanced capabilities that attackers may eventually use.

Governments and AI companies also have a role

The signatories want governments to fund security improvements for hospitals, utilities, local governments, and other organizations that lack cybersecurity resources.

Frontier AI companies such as OpenAI and Anthropic are also expected to provide model access, training, funding, security tools, and direct technical assistance.

The broader objective is to put cyber-capable AI into defenders’ hands before attackers can gain a significant advantage.

AI security has already become a larger focus across the industry. Microsoft recently expanded its AI-driven Windows security strategy, while OpenAI also detailed how it lost control of 1,200 AI agents during a separate security incident.