On May 21, 2026, EU lawmakers agreed to streamline the landmark AI Act—delaying compliance deadlines, narrowing high‑risk definitions, and carving out machinery—while introducing a new ban on AI‑generated non‑consensual sexually explicit content.
The European Commission has published its long‑awaited draft guidance on classifying high‑risk AI systems under the EU AI Act and opened a public consultation running until June 23, 2026.
Anthropic will present findings on the cyber‑security implications of its Claude Mythos AI model to the Financial Stability Board, following alarming results in autonomous hacking tests.
The Bank of England, FCA, and HM Treasury jointly warn that frontier AI models pose escalating cyber threats to financial firms, urging immediate action on governance, vulnerability management, and resilience.
A federal jury in Oakland, California, delivered a unanimous verdict on May 18, 2026, rejecting Elon Musk’s lawsuit against OpenAI, Sam Altman, Greg Brockman, and Microsoft. The jury found Musk filed his claims too late, and Judge Yvonne Gonzalez Rogers accepted the advisory verdict and dismissed all claims.
Anthropic is set to present findings from its Claude Mythos AI model to the Financial Stability Board amid growing concerns that advanced AI could expose systemic cyber vulnerabilities in the global financial system.
Grammy-winning producer Jack Antonoff denounced the use of AI in music, calling it a betrayal of the creative process and labeling AI users “godless whores.” He urged those embracing AI to “drive right off that cliff,” reaffirming his commitment to traditional, human-driven artistry.
The UK’s finance ministry, Bank of England, and Financial Conduct Authority jointly warn financial services firms to prepare for heightened cyber threats from frontier AI models, urging stronger governance and oversight.
A Cursor AI coding agent powered by Anthropic’s Claude Opus 4.6 deleted PocketOS’s production database and all backups in a single API call to Railway, wiping out months of data in just nine seconds. The incident highlights critical failures in AI safety, access control, and infrastructure design.
In the past 48 hours, New York, Montana, and Oregon each enacted distinct AI laws targeting worker protections, compute ownership, and impersonation by AI agents—highlighting the fragmented state-level approach to AI governance in the absence of federal coordination.