OpenAI leads 100 firm call for AI powered cyber defense surge

This is one of the clearest attempts yet by frontier labs and large enterprises to frame AI as a net defensive tool in cybersecurity, not just an offensive force multiplier. By putting OpenAI, Anthropic, Google, Microsoft, Amazon, chipmakers and specialist security vendors on the same open letter, the industry is sending a unified message that status quo cyber hygiene is no longer acceptable in an era of agentic models.

Practically, the letter calls for four concrete shifts: treating cyber defense as a CEO-level priority, arming more defenders with AI tooling, mobilizing vendors to test and harden systems against frontier capabilities, and pushing governments to fund AI-powered protection for under-resourced critical infrastructure. That last piece is critical, because many hospitals, water utilities and local agencies lack in-house talent to safely deploy advanced models.

For the race to AGI, this move does not change capability trajectories, but it shapes how those capabilities are channeled. If even a fraction of the signatories follow through with discounted access, shared tools and joint incident response, we may see a new layer of AI-native security infrastructure become standard. That helps mitigate one of the main tail risks of more capable models: cheap, scalable cyber offense.