Powerful AI models could sharply accelerate cyberattacks in coming months, Five Eyes warns

The Five Eyes intelligence alliance—comprising the United States, the United Kingdom, Canada, Australia, and New Zealand—has issued a warning regarding a rapid surge in cyber threats driven by the development of advanced artificial intelligence models, Euronews reports. 

According to intelligence assessments, new generations of AI are capable of significantly enhancing both defensive and offensive capabilities in cyberspace, with changes expected to manifest not over years, but within the coming months․

AI as an Accelerator of Cyber Threats

The report highlights the dual-use nature of artificial intelligence. On one hand, it helps improve cyber defense by automating vulnerability detection and incident response. On the other hand, it drastically increases the speed, scale, and sophistication of potential attacks by malicious actors. The authors of the document note that future "frontier" AI models could outperform current industry expectations and fundamentally shift the balance between offensive and defensive capabilities in cyberspace. The report’s key phrasing is stark: the timeline is measured in months, not years.

The New Reality of Cybersecurity

Intelligence agencies note that existing AI systems are already capable of discovering and potentially exploiting weaknesses in infrastructure. This applies not only to future models but also to technologies available today. Among the key vulnerabilities that could be amplified by AI, the report lists: weak authentication and access control mechanisms, legacy systems lacking regular updates, inadequate cybersecurity planning within organizations, excessive internet connectivity of critical systems, and slow vulnerability remediation cycles. According to analysts, these specific factors create the foundation for scaling attacks using AI tools.

Shifting the Balance: From Defense to Automated Attacks

The report pays special attention to the fact that as open-source and commercial AI models proliferate, the availability of tools for potential threat actors also increases. The faster their computing power and system analysis capabilities grow, the higher the likelihood of automated attacks, including real-time vulnerability discovery and exploitation. The report emphasizes that traditional cyber risk assessments can become obsolete in literally a matter of months, rather than years.

A Call to Action for Business and Government

Five Eyes urges governments and corporations to act faster and more systematically. Key recommendations include strengthening baseline cyber defenses, tightening access controls, modernizing legacy infrastructure, and integrating cybersecurity into strategic governance. The need to empower security professionals with real authority and resources to respond to emerging threats is heavily emphasized. As the report notes, success will depend not only on technology but also on discipline in its application: "a solid foundation, rapid response, and continuous adaptation to evolving threats."

In Brief

The Five Eyes intelligence alliance warns that advanced artificial intelligence models could drastically amplify cyber threats in the coming months. AI is capable of simultaneously improving defenses and accelerating attacks, making them more large-scale and complex. According to intelligence agencies, the primary risk stems from infrastructure vulnerabilities and legacy systems that can be effectively exploited using new technologies.