U.S. and allied intelligence agencies warn that advanced AI models with dangerous hacking capabilities will become publicly available within months, not years, forcing organizations to shore up basic cybersecurity defenses now.

What happened

The Five Eyes alliance (United States, Canada, UK, Australia, and New Zealand) issued a joint statement warning that frontier AI models like Anthropic's Fable 5 and OpenAI's Daybreak will soon provide advanced hacking capabilities to the public. The agencies stated the timeline is "months," not years, and that AI models capable of exploiting cybersecurity weaknesses are already available today through commercial, open-source, and black-market channels.

What to watch

Programs like Anthropic's Project Glasswing and OpenAI's Trusted Access for Cyber Program are providing AI systems to organizations for cyberdefense, aiming to give defenders a head start. The agencies emphasize that success depends on "getting the basics right, acting quickly, and integrating cyber security into core business strategy"—standard practices that many organizations still treat as afterthoughts.