As Congress advances the reconciliation package, a proposal to enact a 10-year moratorium on state AI laws has emerged as a key consideration. This moratorium has come in response to more than 1,000 AI-related bills introduced by state policymakers in the first few months of 2025 alone. Though well-intentioned, these bills risk creating a fragmented regulatory landscape that could slow innovation and place disproportionate burdens on small businesses developing or adopting AI tools.
Small businesses have already made AI tools an essential part of their operations and would face substantial challenges if access to those tools were limited or complicated by a fragmented regulatory landscape. As we have shared in past testimony, 75 percent of App Association members report using generative AI, and the figure for all types of AI is likely closer to 100 percent. Many rely on internal, fine-tuned models to improve efficiency in areas where they lack expertise, as well as to support essential functions, such as financial management, email marketing automation, and cybersecurity. Consequently, even when they do not develop AI tools themselves, a patchwork of state AI regulations creates downstream effects, such as reduced access to new foundational tools and legal uncertainty, that put everyday uses at risk and damage their bottom line.
While the full financial impact of this multi-state approach to AI regulation remains unclear, early estimates and evidence of disruption to innovation and day-to-day operations warrant delaying additional state laws. To that end, policymakers should seriously consider at least temporarily preempting state laws that specifically seek to regulate AI technologies differently than conduct involving other technologies to provide critical time to collect data, engage with small businesses, and craft an effective national strategy that supports innovation, safeguards competitiveness, and maintains U.S. leadership in AI. Costs to small businesses come in a variety of forms, as summarized below.
A fragmented landscape of state AI laws will impose significant direct and indirect costs on small businesses, from compliance burdens to lost opportunities for growth and innovation. To avoid penalties, small businesses will need to hire legal counsel or divert limited resources toward interpreting a patchwork of inconsistent regulations. For example, the Colorado AI Act, which closely resembles the EU AI Act, imposes penalties of up to $20,000 per violation. If every state and the District of Columbia were to enact similar laws, small businesses could face mountains of fees for potential missteps. Even if some states pursue more permissive, innovation-friendly laws, the cumulative effect of 50 different frameworks would still demand extensive compliance efforts.
Beyond legal costs, these laws could raise operational costs for small businesses and keep them from fully realizing AI-related productivity gains, innovation, and cost savings. Compliance could require devoting limited resources to risk assessments, transparency mandates, record-keeping, and other obligations common in AI regulations. Indeed, in an EU-funded study, researchers found that compliance costs constitute 17.22 percent of the total cost to build an AI system subject to regulation. An increasingly complex regulatory landscape could impose prohibitive and difficult-to-estimate compliance costs on small businesses, undermining their ability to integrate AI into their operations or provide their products and services. Escalating regulatory costs puts these everyday use cases at risk and can also hamper small businesses’ ability to obtain funds from investors who are unlikely to fund compliance programs.
According to a 2023 survey by the Small Business and Entrepreneurship Council, AI saves small businesses $273.5 billion annually. Regulatory uncertainty may deter small businesses’ adoption of AI tools, taking a significant bite out of these savings and productivity gains. Similarly, not knowing whether and to what extent AI systems will be regulated across states will result in a potent disincentive for potential partners and investors. In a high-risk regulatory environment, investors may become more cautious, leaving small tech innovators, many of whom depend on external funding, unable to finance operations. As a result, some may abandon AI projects altogether, delay integrating AI into their operations, or avoid expanding into states with complex or burdensome laws to protect their funding. In fact, a 2024 report by the U.S. Chamber of Commerce found that a quarter of small businesses are currently not using AI-powered tools because of legal or compliance concerns. This chilling effect limits innovation within the AI sector, reduces the availability of affordable AI tools in other industries, such as healthcare, retail, and logistics, and ultimately hurts local job creation and U.S. technological competitiveness.
Compounding these challenges, AI’s rapid evolution will inevitably outpace legislative efforts to regulate it, creating a persistent gap between law and innovation. For example, Apple initially delayed the rollout of AI-powered features, including Apple Intelligence, in Europe due to concerns about compliance with the EU’s Digital Markets Act. Similarly, Meta delayed its multimodal AI models in Europe “due to the unpredictable nature of the European regulatory environment.” As a result, European consumers and small businesses looking to build on foundational AI layers did not have access to tools available elsewhere. States’ current race to regulate AI without coordination or a sufficient understanding of AI’s applications risks creating similar barriers that delay innovation and put U.S. businesses at a competitive disadvantage. A federal moratorium would provide the necessary time to study AI’s development and craft more informed, adaptable policies.
Finally, given AI’s inherently global nature, a fragmented state-by-state regulatory approach would weaken the United States’s position in international AI governance and standards. While the federal government holds the authority to negotiate international agreements, conflicting state laws create uncertainty and undermine the United States’s ability to present a unified position. This lack of cohesion could diminish U.S. influence in international forums, potentially ceding leadership on AI governance to other jurisdictions, such as the European Union.
A patchwork of state AI laws threatens to impose heavy costs on small businesses, slow innovation, and weaken U.S. competitiveness. While the full economic impact remains uncertain, early evidence points to a high risk of significant disruption, particularly for small businesses that lack the resources to navigate 50 different compliance regimes. A federal moratorium on state level AI-regulation would allow time for more research, consultation with small businesses, and the thoughtful consideration of a cohesive national strategy for AI governance. Without this unified approach, policymakers risk trading AI-driven growth and innovation for regulatory confusion and missed opportunities.
