Top Stories – July 28, 2026
1. EU AI Act GPAI Enforcement and Article 50 Transparency Take Effect in Five Days
The EU Commission gains active enforcement powers over general-purpose AI providers on August 2, with fines up to 3% of global annual turnover for violations. Article 50 transparency obligations – including AI content disclosure, chatbot identification, and deepfake labeling – become strictly enforceable across all 27 member states on the same date.
2. EO 14409: Classified Frontier Model Benchmarks Due From NSA, CISA, and NIST on August 1
NSA, CISA, and NIST must deliver classified benchmarks by Thursday to define which AI models qualify as covered frontier models subject to mandatory 30-day pre-release government review under EO 14409. The thresholds these benchmarks set will determine compliance obligations for every major frontier AI lab operating in the United States.
3. Great American AI Act Formally Introduced in the House
Reps. Jay Obernolte (R-CA) and Lori Trahan (D-MA) formally introduced a House companion bill on July 23 covering frontier AI auditing, transparency reporting, and whistleblower protections. The Senate passed its version of the bill 67-31 on July 3, setting up a potential bicameral path to the first comprehensive federal AI law.