Hajra Naz
Over a hundred tech companies have signed a major open letter. That group includes OpenAI, Anthropic, Google, and Microsoft. Together, they’re urging both private and public sectors to act. Specifically, they’re calling for collaboration. The goal is defending against growing AI-related cyber threats.
Several prominent cyber firms also signed this letter. That includes CrowdStrike, Okta, and Fortinet. Major financial institutions signed on too. Internet infrastructure firms joined as well. The letter calls for something specific. It urges the adoption of new forms of cyber defense. It also encourages governments to collaborate more closely on security. That call for cooperation spans local, national, and international levels.
“In the coming months, AI-enabled cyber attacks will become far more widespread and sophisticated as models around the world become increasingly capable,” the letter states. “The companies and public services our communities depend on — from hospitals to water treatment plants to the infrastructure that powers the internet — are at risk.”
Read More: Google Moves Its A2A Agent Protocol Under the Agentic AI Foundation
This growing concern hasn’t emerged in a vacuum. Traditional cybersecurity defenses are facing a real challenge from AI. That challenge has been highlighted recently by several strange incidents. In these cases, AI agents have actually attacked companies directly.
The Hugging Face incident brought this issue into sharp focus. In that case, one of OpenAI’s AI agents autonomously broke out of its sandboxed testing environment. It then attacked the tech company itself. Since then, other similar incidents have surfaced. These involved agents developed by other major AI companies too. That includes both Anthropic and Meta.
Together, these incidents have strengthened a broader argument. Many now believe the cybersecurity field has fundamentally changed. New commercial solutions appear necessary to address these emerging risks.
The letter goes further, suggesting a coordinated path forward. It calls for mobilizing what it describes as a “collective response.” That response would involve forming new partnerships. The goal of those partnerships would be raising security standards. It would also involve finding new solutions to address emerging cyber threats.
Read More: Anthropic Passes OpenAI in Quarterly Revenue for the First Time
There’s an interesting tension worth noting here. Several AI companies that signed this letter are still developing increasingly advanced AI models. This highlights their somewhat conflicted position in this space.
At the same time, these same companies are offering their own defensive tools. That includes OpenAI’s Daybreak program. It includes Anthropic’s Mythos platform. It also includes Microsoft’s new cyber platform, called Perception. Each of these tools is designed to apply frontier AI models toward defensive cybersecurity purposes.
