A Republican-led House committee has proposed a budget reconciliation bill aimed at prohibiting states from regulating artificial intelligence. Introduced on Sunday, the legislation seeks to prevent states from enforcing any laws related to a wide variety of automated computing systems for a decade following its enactment. This legislation could hinder efforts to regulate technologies ranging from AI chatbots to online search algorithms.
Critics, including Democrats and organizations advocating for responsible AI development, have expressed concerns that the bill represents a significant advantage for major tech firms. Advocacy groups such as Americans for Responsible Innovation (ARI) warn that the proposed measures could have dire implications for public welfare. This comes at a time when companies like OpenAI are actively seeking to alleviate pressures from a range of state regulations.
The legislation, spearheaded by House Energy and Commerce Committee Chairman Brett Guthrie (R-KY), aims to eliminate “legal impediments” for AI models and automated decision-making systems. It categorizes these systems as computational processes that utilize machine learning and other forms of data analytics to provide simplified outputs that can significantly influence or replace human decision-making.
This ten-year ban may have extensive implications beyond just AI, according to Travis Hall, the director for state engagement at the Center for Democracy & Technology. He noted that the automated decision systems mentioned in the bill are widely integrated into digital services including search engines, navigation tools, and even healthcare diagnostics.
During the forthcoming 2025 legislative session, states have already proposed more than 500 bills that could be unequivocally obstructed by this legislation. These proposed laws address a range of issues, from ensuring chatbot safety for minors to imposing restrictions on deepfakes and mandating disclosures regarding AI usage in political advertising. Should the bill pass, states that have already enacted their own AI regulations might find their initiatives rendered ineffective.
“The move to ban AI safeguards is a giveaway to Big Tech that will come back to bite us.”
In recent developments, California Governor Gavin Newsom enacted a law preventing the use of a performer’s AI-generated likeness without consent. Meanwhile, Tennessee and Utah have passed legislation offering similar protections, and Colorado is set to enforce a new law next year requiring developers of high-risk AI systems to help protect users from algorithmic discrimination.
California nearly passed a significant AI safety law, SB 1047, which aimed to impose stringent security requirements and legal liabilities on AI companies. OpenAI opposed the measure, advocating instead for a unified federal approach to prevent a fragmented patchwork of state regulations that could hinder innovation. Ultimately, Governor Newsom vetoed the bill last September, further indicating the company’s desire to avoid state-level restrictions on its operations.
With minimal regulation around AI at the federal level, individual states have taken the lead in addressing the implications of this technology. Even prior to the rise of generative AI, state lawmakers were grappling with issues surrounding algorithmic discrimination, including biases displayed by machine learning systems in sectors like housing and criminal justice. Efforts to address these issues may be greatly impacted by the new Republican proposal.
Democratic representatives have criticized the inclusion of this provision, with Rep. Jan Schakowsky (D-IL) asserting that the proposed 10-year ban would result in AI companies neglecting consumer privacy protections and spreading harmful technologies like deepfakes. Senator Ed Markey (D-MA) echoed these concerns, expressing in a statement that the proposal would lead to dire consequences for the environment and vulnerable communities.
The nonprofit ARI has drawn parallels between this potential ban and past governmental failures to regulate social media, highlighting the long-term consequences of inaction. ARI President Brad Carson remarked that lawmakers’ hesitance to implement social media safeguards has had lasting negative effects. He cautioned that similar failures in regulating AI technology could pose even greater risks due to its rapid evolution.
The path for the proposal in the Senate remains uncertain, as the ARI points out that the Byrd Rule restricts reconciliation bills to fiscal matters. Nonetheless, the push by Republican lawmakers to block state-level oversight raises concerns about the future of AI governance.
