Overview
Severity: MEDIUM | Affected: US AI Industry | Category: policy
The United States Congress has passed the landmark 'AI Model Accountability Act' (AMAA), a bipartisan bill aimed at increasing transparency and security for powerful AI systems. The new law mandates that companies developing or deploying 'high-risk' AI systems—defined as models with over 10^25 FLOPs of training compute or those used in critical infrastructure sectors—must undergo regular, independent third-party security audits. These audits will assess vulnerabilities to adversarial attacks, data poisoning, and potential for misuse. The act also establishes a public registry for these high-risk models, detailing their training data sources and safety mitigation measures. Proponents hail the legislation as a crucial step towards responsible AI governance, while some industry groups have raised concerns about the compliance burden and the potential for stifling innovation. The bill is expected to be signed into law by the President next month.