Five Eyes intelligence alliance warns AI models capable of taking down governments are months away

In a rare joint statement, the Five Eyes intelligence alliance warns that advanced AI models capable of taking down governments and businesses are mere months away. The signals agencies of Australia, the US, the UK, New Zealand, and Canada urge leaders to act immediately. This warning follows the Trump administration's June 2026 decision to block foreign nationals from accessing Anthropic's powerful Fable and Mythos models, which are capable of identifying critical cyber vulnerabilities.

Five Eyes cybersecurity warning

• ▪The Five Eyes alliance comprises the signals intelligence agencies of Australia, the United States, the United Kingdom, New Zealand, and Canada
• ▪The Five Eyes intelligence alliance issued a joint public statement warning that advanced AI models capable of taking down governments and businesses are months away
• ▪The Five Eyes agencies urged political and business leaders to act immediately, stating that cyber risk must be treated as a core business risk and leadership responsibility

Anthropic Mythos model capabilities

• ▪Anthropic's Mythos is a powerful AI model capable of detecting vulnerabilities in cyber systems and is restricted to vetted organizations and companies
• ▪Anthropic employees are reportedly working with the National Security Agency, and the Mythos model is reportedly powering offensive cyber operations against China and Iran

Anthropic Fable model restrictions

• ▪Anthropic's Fable 5 is a more community-friendly variant of the Mythos model designed to detect vulnerabilities in digital infrastructure

US foreign national access ban

• ▪The Donald Trump administration blocked foreign nationals from using Anthropic's Fable and Mythos AI models in June 2026, citing national security advice
• ▪United States intelligence agencies had early access to Anthropic's Fable 5 and Mythos 5 models prior to the restriction on foreign nationals

Australia national AI strategy

• ▪Australia's national AI plan promotes a light-touch regulatory approach to capture economic and productivity benefits while voluntarily promoting safety
• ▪The Australian government signed a non-binding memorandum of understanding with Anthropic in March 2026 as the first company on its national AI plan

AI cyber threat timeline

• ▪The Five Eyes alliance warned that frontier AI models will accelerate the speed, scale, and sophistication of cyber threats
• ▪National security expert Olivia Shen stated that other advanced AI models, potentially developed by states like China, may be nearing release