Five Eyes Warns Frontier AI Could Transform Cyber Threats Within Months

Five Eyes Warns Frontier AI Models Could Rapidly Alter Cybersecurity Risks

"Cyber risk is a core business risk and a leadership responsibility"

Reuters says the warning signals rising concern among intelligence agencies over models such as Anthropic's Mythos and OpenAI's GPT 5.5-Cyber

The Five Eyes intelligence alliance of the US, UK, Canada, Australia and New Zealand warned that cybersecurity threats stemming from artificial intelligence could become reality within months and called for immediate action.

In a joint statement released on July 22, the alliance said cutting-edge frontier AI models could fundamentally change cyber capabilities on both offense and defense beyond current industry expectations. The technological shift could come in months, not years, it said.

The group urged national leaders to act now. It called for assessments of AI-related risks, readiness and accountability; prioritizing basic cybersecurity practices and control systems; giving cyber leaders full authority and resources; and maintaining active engagement with evolving threat indicators and guidance.

The intelligence agencies also said organizations and society as a whole need to respond. Cyber risk is no longer just a technical issue but a core business risk and a leadership responsibility, the statement said.

The statement did not cite specific examples or evidence of AI-based cyberattacks.

In the market, the warning is seen as an extension of broader security concerns that have persisted since the so-called Mythos shock, which raised fears that AI could fundamentally disrupt the existing security ecosystem.

Still, the warning carries added weight because intelligence agencies directly addressed cyber threats linked to AI.

Reuters said the warning suggests global intelligence authorities' concerns over advanced AI models such as Anthropic's Mythos and OpenAI's GPT 5.5-Cyber have reached a critical level.

No specific model was named in the statement. But it is widely viewed as a reference to Anthropic's Mythos, which has recently been hit by export controls.

Earlier, on July 12, Anthropic said the US government issued export-control guidance fully blocking foreign access to Claude Mythos 5 and Claude Fable 5.

South Korea's government and some companies recently joined Project Glasswing, which offers access to Mythos. But the latest action by the US administration has effectively constrained use of the model.

Park Sang-kyung, Hankyung.com reporter highseoul@hankyung.com