States take the lead in AI regulation as federal government steers clear

As the federal government remains hesitant to implement comprehensive regulations on artificial intelligence, individual U.S. states are taking the initiative to establish crucial guidelines surrounding AI technologies. A recent setback in Congress, where a proposed nationwide moratorium on state-level AI regulation was rejected, allows states to continue crafting their own regulations. In 2025 alone, all 50 states have introduced various pieces of legislation aimed at regulating AI, focusing on critical areas such as government usage, healthcare applications, facial recognition technologies, and generative AI. The need for oversight in these areas is increasingly important, particularly within the public sector, where predictive AI is reshaping government functions ranging from determining eligibility for social services to influencing decisions in the criminal justice system. However, the rise of algorithmic decision-making raises concerns over potential hidden consequences. Issues such as racial and gender biases are of particular concern, prompting state legislatures to propose bills that emphasize transparency, consumer protections, and the need to understand the risks associated with AI deployment. Several states have mandated that AI developers disclose the risks linked to their technologies. For instance, Colorado’s Artificial Intelligence Act implements transparency and disclosure requirements for developers involved in significant decision-making processes. Meanwhile, Montana has introduced a “Right to Compute” law that requires AI developers to adopt risk management frameworks aimed at addressing security and privacy issues in critical infrastructure development. Furthermore, states like New York are establishing regulatory bodies to oversee AI practices, as highlighted in their SB 8755 bill.