Pritzker signs landmark AI regulation bill that aims to mitigate risks

What Happened

Illinois Governor J.B. Pritzker has signed Senate Bill 315, the Artificial Intelligence Safety Measures Act, into law, adding the state to a small but growing list of U.S. jurisdictions moving to regulate advanced artificial intelligence systems directly. The law targets only the largest AI developers — those whose models generate more than $500 million in annual revenue and are trained using very large amounts of computing power — and imposes new transparency and accountability requirements on them. Smaller AI companies and startups appear to fall outside the law's scope, at least for now.

Why It Targets the Biggest Players

By setting a high revenue and compute threshold, Illinois is following a pattern seen in other so-called "frontier model" regulations: rather than regulating all AI uniformly, lawmakers are trying to focus oversight on the handful of companies capable of building the most powerful, potentially highest-risk systems — think OpenAI, Google, Anthropic, and Meta rather than smaller AI startups. This approach mirrors debates happening at the federal level and in California, where similar frontier-model safety bills have been proposed or passed, and reflects a broader belief among regulators that the biggest risks — from mass disinformation to safety failures in high-stakes applications — scale with model size and resources.

Why It Matters Nationally

With Congress still gridlocked on comprehensive AI legislation, states have become the primary battleground for AI governance in the U.S. Illinois joins California, Colorado, and a handful of other states that have passed or advanced AI-specific laws in the past two years. This patchwork approach creates real challenges for AI companies operating nationally, who now must track a growing list of state-level transparency, disclosure, and accountability rules that don't always align. It also raises the perennial question of whether Congress will eventually preempt these state laws with a unified federal framework — an outcome the tech industry has lobbied heavily for, arguing that fragmented state rules stifle innovation.

The EU AI Act Comparison

Illinois' law is far narrower than the European Union's AI Act, which imposes tiered obligations across a much broader range of AI systems, including risk classifications for everything from biometric surveillance to hiring algorithms. Still, the underlying philosophy — that the most powerful and consequential systems deserve the most scrutiny — echoes the EU's approach to "systemic risk" models. As U.S. states experiment with narrower, sector-specific rules, the comparison to the EU's more comprehensive regime highlights an ongoing divergence in regulatory philosophy: the U.S. moving piecemeal through state action, while the EU pursues a single, sweeping statute.

What to Watch

Expect legal challenges, industry pushback, and pressure for federal preemption as more states craft their own AI safety measures, each testing different thresholds, definitions, and enforcement mechanisms.