Introduction
In 2024 [1] [2] [3], there was a significant surge in legislative activity concerning artificial intelligence (AI) at the state level in the United States. Nearly 700 AI-related bills were introduced across 41 states [1], focusing on private AI applications, safety standards [1], deepfake regulations [1], and digital content watermarking [1]. This legislative push highlights a growing effort to address regulatory gaps left by stalled federal action.
Description
In 2024 [1] [2] [3], significant legislative activity regarding artificial intelligence (AI) occurred at the state level [3], with nearly 700 AI-related bills introduced across 41 states [1], marking a notable increase from the previous year. The primary focus was on private uses of AI [3], safety standards [1], deepfake regulations [1], and digital content watermarking [1], with California [1] [2] [3], Colorado [1] [2] [3], Utah [3], and Illinois leading the way [3]. Key legislative actions also addressed issues such as non-consensual sexual imagery, political deepfakes [2], and copyright protections [2], reflecting a regulatory gap due to stalled federal action [1].
California’s AB 2013 mandates that developers of generative AI systems disclose training data on their websites for models released after 2021 [3]. AB 942 requires marking AI-generated content and providing a free detection tool [3], while AB 1008 expands the definition of personal information under the California Consumer Privacy Act to include data from AI systems [3]. Notably, SB 1047 [1] [3], which aimed to enforce safety protocols for large AI models [3], was vetoed by the governor [3], emphasizing the need for more refined proposals [1].
Colorado enacted SB 205 [3], the first comprehensive AI law in the US [3], targeting discrimination in high-risk AI systems that make consequential decisions in areas like employment and healthcare [3]. This law imposes transparency and accountability requirements on companies [3], although broader AI legislation faced challenges in states like California and Connecticut [2]. Concerns have been raised about the potential for overly broad state regulations to hinder technological advancement [1], highlighting the importance of balancing regulation to avoid stifling innovation [1].
Utah’s SB 149 emphasizes transparency [3], requiring notification to consumers when they interact with AI systems [3]. Illinois’ HB 3773 mandates that employers using AI for employment decisions must notify employees and prohibits discrimination based on protected classes [3]. The assignment of liability among AI developers [1], deployers [1] [3], and users remains a critical issue as states continue to advance their own legislation.
At the federal level [3], the FTC intensified enforcement against deceptive AI practices [3], including actions against companies making false claims about AI tools [3]. The agency also required the destruction of models trained on improperly collected data [3]. The Texas Attorney General settled with a tech company over misleading statements regarding an AI tool for healthcare [3].
Federal agencies are implementing measures from President Biden’s Executive Order on AI [3], focusing on prohibiting high-risk activities [3], such as social scoring and untargeted facial recognition scraping [3], starting in February 2025 [3]. General Purpose AI Model providers will face new documentation and compliance requirements beginning in August 2025 [3], with additional obligations for high-risk systems [3]. The Biden administration has emphasized safety protocols [1], while the Trump campaign’s stance on AI regulation reflects both criticism of big tech and opposition to regulation [1].
Privacy laws remain primarily state-driven [3], with Congress failing to pass a comprehensive federal privacy bill [3]. However, the Kids Online Safety and Privacy Act is gaining bipartisan support [3]. California and Colorado have updated their definitions of personal data to include neural data and biometric protections [3]. The lack of unified federal standards complicates the establishment of effective AI oversight frameworks [1].
California has initiated rulemaking on several AI-related topics [3], with significant projected costs for compliance [3]. Regulatory scrutiny in the privacy space has intensified [3], particularly concerning sales [3], minors’ data [3], and location information [3]. The Department of Health and Human Services has also ramped up enforcement actions [3], particularly related to breaches of privacy and security rules [3].
Overall, the landscape for AI legislation and regulation is rapidly evolving [3], with expectations for increased activity in 2025 [3], influenced by the incoming administration’s tech policy agenda [2]. Democratic-led states are anticipated to take the lead in regulating areas such as AI and cryptocurrency [2], particularly in response to federal deregulation efforts [2], while Republican-led states may focus on promoting AI innovation and deregulating cryptocurrency [2]. The energy demands of AI technology pose immediate challenges [1], with data centers consuming significant power [1], and deepfake technology is highlighted as a threat to democratic stability [1], prompting calls for educational initiatives to better understand its implications [1]. In healthcare [1] [3], overregulation of AI could stifle innovations [1], with recommendations for targeted oversight that prioritizes patient safety and leveraging existing regulatory frameworks rather than creating new bodies [1]. Key recommendations include providing pathways for updated approvals and ensuring regulations align with existing protections under laws like HIPAA [1].
Conclusion
The legislative landscape for AI in the United States is rapidly evolving, with states taking the lead in addressing regulatory gaps left by federal inaction. This surge in state-level legislation reflects a growing recognition of the need to balance innovation with regulation to ensure safety and accountability in AI applications. As the federal government and states continue to navigate this complex terrain, the implications for technological advancement [1], privacy [3], and democratic stability remain significant. The ongoing efforts to establish effective oversight frameworks will be crucial in shaping the future of AI regulation and its impact on society.
References
[1] https://www.pymnts.com/artificial-intelligence-2/2024/states-lead-ai-regulation-push-as-us-policy-interests-shift-with-new-administration/
[2] https://csmapnyu.org/impact/news/new-report-state-lawmakers-pass-more-than-200-tech-policy-laws-in-2024
[3] https://www.jdsupra.com/legalnews/year-in-review-privacy-compliance-and-1831363/
