Top Stories - September 16, 2026
Brazil Chamber of Deputies to Vote on Latin America’s First Comprehensive AI Law
Brazil’s Chamber of Deputies is scheduled for a plenary floor vote today on Bill 2338/2023, a risk-based framework already passed by the Senate that would make Brazil home to Latin America’s first comprehensive AI law, with high-risk designations covering biometric identification, judicial support systems, credit evaluation, and critical infrastructure AI.
EU AI Act GPAI: First Systemic Risk Evaluation Deadline Passes
September 15 marked the first binding enforcement milestone under the EU AI Act’s general-purpose AI obligations - providers of models trained above the 10^25 FLOPs threshold were required to submit formal systemic risk evaluations to the European AI Office covering red-teaming results, energy consumption disclosures, and copyright training-data compliance.
China Releases AI Safety Governance Framework 3.0, Targeting Agents and Control Risks
China’s TC260 cybersecurity committee released version 3.0 of its AI Safety Governance Framework on September 14 during National Cybersecurity Week, adding a dedicated agent-risk annex addressing autonomous tool use, embodied AI, and loss-of-human-control risks - TC260 frameworks historically shape China’s technical standards and regulatory enforcement expectations.