- China claims to have developed an AI 'cyber nuclear weapon' that rivals advanced US models, signaling a new era of cyber warfare.
- The U.S. company Anthropic's 'superhuman' AI model, Mythos, can identify and exploit sophisticated software vulnerabilities, prompting a U.S. government directive to restrict its access.
- These advanced AI models can discover and weaponize software vulnerabilities at an unprecedented scale, threatening critical infrastructure such as power grids and financial networks.
- Experts warn that an AI-powered cyber war could cause widespread societal disruption, with human oversight potentially hindered by machine speed.
- The Five Eyes intelligence agencies have issued a rare joint statement, emphasizing that frontier AI models pose the greatest risk to critical computing systems within months and urging immediate action.
IN FULL
