Five Eyes intelligence alliance warns AI could enable cyber attacks on governments

In a rare joint statement released on 22 June, the Five Eyes intelligence community—comprising Australia, the United States, the United Kingdom, Canada and New Zealand—warned that advanced artificial‑intelligence models are developing faster than anticipated and could be exploited by malicious actors to paralyse governments, critical infrastructure and businesses.

The alliance said AI lowers the barriers for attackers, speeds up exploitation and shortens the window between discovery and exploitation, noting that “time is measured in months, not years” and that “cyber risk is no longer a purely technical issue but a core business risk.” It urged organisations to strengthen digital defences, keep software up‑to‑date, restrict privileged access and prepare for attacks before they occur. The statement referenced recent controversy over US AI startup Anthropic’s models Mythos and Fable 5, which were restricted after US officials barred foreign citizens from using them over national‑security concerns.