AI Regulation in the US: Where Things Stand in 2026

AI regulation in the United States looks very different from Europe. There is still no single, comprehensive federal AI law in October 2026. Instead, businesses face a mix of federal agency enforcement under existing laws, a growing body of state AI legislation, and an active federal effort to limit state rules. This guide explains where things stand, the main types of state laws, and what organisations and creators who use AI should do. It is general information, not legal advice, and rules are changing quickly.

The federal picture

No comprehensive AI statute

Congress has debated many AI bills but has not passed a broad law covering AI across sectors. Proposals to block states from regulating AI for a period of years were considered in 2025 but not enacted.

Executive action

Federal AI policy has shifted with each administration. In December 2025, the White House issued an executive order titled "Ensuring a National Policy Framework for Artificial Intelligence". It directs federal agencies to evaluate state AI laws, sets up a task force to challenge laws the administration considers onerous, and asks for legislative recommendations for a uniform federal standard. An executive order does not by itself cancel state laws; whether particular state rules are preempted is likely to be decided by courts or Congress.

Existing laws still apply

Federal agencies continue to apply long-standing laws to AI. Consumer protection law prohibits unfair or deceptive practices, including exaggerated claims about what an AI product can do. Anti-discrimination laws in employment, housing and credit apply regardless of whether a decision involves AI. Sector regulators, such as those for medical devices and finance, also oversee AI in their areas.

Voluntary frameworks

Many organisations use the NIST AI Risk Management Framework as a practical guide to identifying and managing AI risks. It is voluntary but widely referenced, including in some state laws.

State AI laws: the main categories

States have been the most active AI lawmakers. Law-firm trackers and legislative databases report that nearly every state introduced AI bills in 2025, with dozens enacting at least one. The laws fall into a few broad groups:

• Comprehensive high-risk AI laws: Colorado's AI Act is the best-known example, requiring developers and deployers of "high-risk" AI systems to take reasonable care to prevent algorithmic discrimination, carry out impact assessments and give consumers notices. Its start date has been pushed back and its details revised more than once, so check the current status.
• Transparency and chatbot disclosure: several states require businesses to tell people when they are interacting with AI in certain contexts, and some have rules specific to companion chatbots and minors.
• Frontier model safety: California has passed laws aimed at developers of the most powerful models, focused on safety frameworks, incident reporting and transparency about training data.
• Employment: Illinois was an early mover, regulating AI analysis of video job interviews, and several states and cities regulate automated tools in hiring.
• Healthcare: some states restrict AI as the sole basis for insurance coverage denials or require disclosure when AI writes patient communications.
• Deepfakes: many states ban deceptive AI-generated media in elections and non-consensual intimate deepfakes.

Privacy laws and AI

Comprehensive state privacy laws, led by California's, give consumers rights over personal data and increasingly address automated decision-making. California's privacy regulator, the California Privacy Protection Agency, has adopted rules on automated decision-making technology, risk assessments and cybersecurity audits. If your AI tools process personal information about residents of these states, privacy law is often the first place obligations arise.

What US rules mean for non-US businesses

State laws generally apply based on where consumers are, not where a company is based. A business in Nigeria or Europe that offers AI-powered services to US residents may need to consider them, especially laws on privacy, chatbots and deepfakes. Our overview of AI regulation around the world compares the US approach with the EU and Nigeria.

Practical steps for organisations

1. Map your AI uses. List where you use AI and whether any use affects decisions about people, such as hiring, credit or housing.
2. Avoid overclaiming. Describe AI features accurately in marketing.
3. Disclose AI interactions. Make it clear when customers are dealing with a bot.
4. Protect personal data. Follow applicable privacy laws and limit what you share with AI tools.
5. Keep humans involved. Have a person review consequential AI-supported decisions.
6. Track changes. The federal-state situation is moving; review your obligations regularly.

Low-risk uses, such as drafting blog posts with the AI Writer or brainstorming in AI Chat, rarely trigger specific AI law, but general rules on accuracy, copyright and privacy still apply.

Because this area changes month to month, review official sources and your legal adviser's guidance before relying on any specific rule.

Key takeaways

• The US has no comprehensive federal AI law as of October 2026.
• State laws create a patchwork of obligations.
• A December 2025 executive order targets state AI laws but does not repeal them.
• Existing consumer protection and anti-discrimination laws still apply.

Related guides

Keep reading in this topic cluster:

• Data Privacy and AI Tools: What Happens to Your Data
• Who Is Responsible When AI Makes a Mistake?

Frequently asked questions

Is there a federal AI law in the US?

Not a comprehensive one as of October 2026. Existing federal laws apply to AI, and many states have their own AI laws.

Can the federal government override state AI laws?

Congress can preempt state laws by statute. An executive order can direct agencies and litigation but does not on its own repeal state law.

Which states have the most AI laws?

Trackers commonly point to California, Colorado, Utah, Texas, Illinois and New York as among the most active.

Does US AI law apply to small businesses?

Often yes, depending on the law, the use case and the business's size and customers. Some laws include thresholds or exemptions.

Where can I track new US AI laws?

Legislative databases maintained by state-legislature organisations and law-firm trackers are updated regularly. Always confirm details against the official text of the law or with a qualified lawyer before relying on them.