Regulating the Frontier: How California, New York, and Illinois Are Governing the Most Powerful AI Models | American Enterprise Institute - AEI

The American Enterprise Institute published a report titled "Regulating the Frontier: How California, New York, and Illinois Are Governing the Most Powerful AI Models." According to the report, California, New York and Illinois have enacted frontier-AI oversight legislation addressing transparency, safety frameworks and incident reporting. The title names the three states and the subject under examination: the governance of the most powerful AI models. The report states that the three states have put oversight measures into statute.

The report argues that mandated transparency is the right starting point for frontier-AI oversight. It states that transparency mandates hold developers to their safety commitments without "putting a regulator between a model and its release." The report's framing distinguishes mandated disclosure from a pre-release approval or licensing process in which a regulatory body would review a model before deployment. The summary of the report does not enumerate the specific obligations, thresholds, penalties or effective dates contained in each state's legislation.

The report is published by the American Enterprise Institute and is limited in scope to three states: California, New York and Illinois. It addresses frontier models specifically, described in the title as "the most powerful AI models." The report characterises the enacted legislation as covering three areas — transparency, safety frameworks and incident reporting — and presents transparency as the appropriate first element of frontier-model governance. Its conclusion concerns the design of oversight rather than its absence, endorsing mandated transparency as a mechanism for accountability while arguing against a pre-release regulatory step.