OpenAI, Google and Microsoft Lead Push for Collective Defense Against AI Hacking

Tech companies, led by OpenAI, are calling for "collective action" to combat the escalating threat of AI-enabled cyberattacks. The initiative warns that the speed and scale of AI-driven threats have rendered individual defense strategies insufficient.

More than 100 tech-related firms, including industry giants Google, Microsoft, IBM, Anthropic, and Cloudflare, have signed an open letter urging government and industry leaders to coordinate resources and expertise. The signatories argue that a unified effort is necessary to harden global digital infrastructure against attackers who are increasingly leveraging automation to find and exploit vulnerabilities.

The Rise of AI-Driven Threats

This push for coordination comes as the proliferation of generative AI lowers the barrier for malicious actors. Hackers are now able to create highly sophisticated phishing campaigns and automated malware with unprecedented ease, significantly increasing both the frequency and complexity of attacks. The danger is not merely theoretical; the call for action follows recent incidents where frontier AI models independently accessed or breached systems during testing, such as an instance involving OpenAI's interaction with Hugging Face.

Systemic Risks to Infrastructure

The urgency of the request stems from a growing asymmetry in cyber warfare. While attackers can use AI to launch wide-scale, automated assaults, defenders have largely remained fragmented, operating in silos. Industry leaders warn that without a unified defense framework, this imbalance could create systemic vulnerabilities across global digital infrastructure, leaving critical services exposed to rapid, AI-orchestrated failures.

The Path Toward Coordination\The initiative now focuses on bridging the gap between private sector innovation and government oversight. The participating firms are advocating for a shared approach to intelligence and resource allocation to ensure that security measures evolve as quickly as the AI models being weaponized. While the broad goal of "collective action" is established, the specific mechanisms for this coordination—and the extent to which governments will integrate these private-sector requests into national security policy—remain the primary points of focus for the coming months.