U.S. Faces 1,200 AI Bills Without Clear Regulatory Standards

The United States is grappling with a fragmented and rapidly expanding landscape of artificial intelligence legislation, having introduced over 1,200 AI-related bills in 2025 alone. Experts Jeffrey Sonnenfeld, Stephen Henriques, and Gary Marcus argue that policymakers lack a unified framework to distinguish necessary regulations from legislative noise. This regulatory chaos spans state-level initiatives, such as California’s SB 53 and New York’s RAISE Act, which impose varying compliance burdens on tech companies. Simultaneously, federal policy remains inconsistent, with conflicting executive orders and defense acts failing to establish a coherent national standard. IBM CEO Arvind Krishna warns that excessive bureaucracy could hinder U.S. competitiveness in the global AI race against the EU and China. The recent disclosure of Anthropic’s powerful Mythos Preview model has further intensified calls for an FDA-like vetting system for advanced AI. Without a shared test for effective policy, the U.S. risks creating a bloated regulatory environment that stifles innovation while failing to address critical security risks, leaving businesses and lawmakers without clear guidance in an increasingly competitive international technological arena.