The EU AI Act creates a binding, risk-based framework for artificial intelligence, while the United States has a more fragmented approach built from federal agency authority, state laws, and sector-specific rules. For businesses, the practical difference is that EU obligations can apply to a product or service supplied in Europe, whereas US requirements depend more on where and how the system is used, which laws and agencies apply, and whether state rules are triggered.
How the two approaches developed
The EU adopted the AI Act as Regulation (EU) 2024/1689. It entered into force on August 1, 2024, with obligations phased in over time. It applies across the EU and is designed to harmonize requirements for AI systems and general-purpose AI models. Its central organizing principle is risk: prohibited practices face bans, certain uses are classed as high-risk, and some systems have transparency obligations.
The United States does not have one comprehensive federal AI statute equivalent to the EU AI Act. Federal agencies apply existing laws within their remits—for example, rules concerning consumer protection, discrimination, safety, privacy, or financial services may be relevant when AI is involved. Federal executive policies can guide government activity and agency priorities, but they are not the same thing as a general, enacted AI statute applying to every private business.
States and local governments add another layer. Some have enacted laws addressing particular AI uses, such as automated employment decisions or consumer-facing systems; others impose disclosure, assessment, or governance duties in defined circumstances. Requirements and effective dates vary, and legislation changes. Businesses should check current law for each place where a system is developed, offered, or used.
Scope and who carries responsibility
The EU Act can reach companies outside the EU. In broad terms, it covers providers placing AI systems or general-purpose AI models on the EU market, providers and deployers established in the EU, and certain non-EU providers or deployers when an AI system’s output is used in the EU. A company cannot assume that operating from the United States alone removes EU obligations.
The Act distinguishes roles. A provider develops an AI system or has one developed and markets it or puts it into service under its name. A deployer uses an AI system under its authority, other than for a personal, non-professional activity. Importers, distributors, and product manufacturers may also have duties. A business that modifies a system or changes its intended purpose can, in specified situations, take on provider responsibilities.
In the US, responsibility is less often assigned through one cross-sector AI statute. A company may instead be accountable under existing rules as an employer, lender, seller, platform, healthcare organization, or data controller, depending on its activity. A vendor contract does not necessarily eliminate the customer’s legal responsibility, and buying a model from a supplier does not automatically make the buyer a mere bystander.
Risk categories and the rules they trigger
The EU framework identifies prohibited AI practices, high-risk systems, systems subject to transparency duties, and other systems that generally face fewer AI-Act-specific requirements. The classification depends on the system’s intended purpose and context, not simply on whether it uses advanced technology. Some uses in areas such as employment, education, essential services, law enforcement, and migration may fall into high-risk categories, subject to the Act’s detailed definitions and exceptions.
- Prohibited practices: Certain specified uses are banned, subject to the wording and exceptions in the Act.
- High-risk systems: Providers face requirements that include risk management, data governance, technical documentation, record-keeping, transparency to deployers, human oversight, and accuracy, robustness, and cybersecurity measures. Conformity assessment and other duties may apply before market placement.
- Transparency cases: In defined situations, people must be informed that they are interacting with AI or that content has been generated or manipulated by AI. The exact duty depends on the use case.
- General-purpose AI models: The Act sets separate obligations for providers, with additional requirements for models presenting systemic risk.
There is no single matching US risk ladder for all businesses. Instead, the legal test may concern the outcome or practice: for example, whether an employment tool unlawfully discriminates, a marketing claim misleads consumers, or a financial decision violates applicable requirements. A state law may impose an AI-specific assessment or notice duty for a particular use, even when federal law does not prescribe a general AI classification.
What compliance looks like in practice
An EU-facing business should first map its systems, intended purposes, users, and role in the supply chain. It should then determine whether an AI Act category applies and identify the corresponding duties. For a high-risk system, this can mean maintaining documentation, implementing a quality-management process, monitoring risks, keeping records, and providing information that enables appropriate human oversight. Deployers have their own responsibilities; they should not assume that the provider’s conformity work covers every operational duty.
US compliance usually starts with a use-case and jurisdiction review. Identify the existing laws relevant to the activity, the agencies that regulate the business, and any state or local rules covering the system. For example, a company using automated tools to screen job applicants may need to examine employment-discrimination rules and any applicable state or local requirements, while a bank using AI for lending must consider the rules governing credit decisions. The precise obligations depend on facts, not just on the label “AI.”
In both regions, practical controls often include an inventory of AI tools, vendor due diligence, testing for foreseeable failures and unfair outcomes, clear notices where required, access controls, incident handling, and records showing how decisions are made. These are useful governance measures, but they do not by themselves prove legal compliance.
Cross-border planning and common misconceptions
A company serving both markets may need to meet overlapping requirements rather than choose one regime. A system might be high-risk under the EU Act and also fall under US employment, consumer, privacy, or sector-specific rules. The stricter operational requirement may be a sensible internal baseline, but legal obligations still need to be assessed jurisdiction by jurisdiction.
- “The US has no AI rules” is inaccurate. It lacks one comprehensive federal AI statute, but existing federal laws, agency enforcement, and state or local measures can apply.
- “Only EU companies need to follow the AI Act” is inaccurate. Certain non-EU providers and deployers can be in scope when their systems or outputs are connected to the EU market or users.
- “Every AI system is high-risk” is inaccurate. The Act’s classification depends on defined categories and intended use; many systems are not high-risk, though other duties may still apply.
- “A vendor handles all compliance” is unsafe as an assumption. Provider and deployer responsibilities can differ, and contracts cannot necessarily transfer statutory duties.
Keep a change process in place. A model update, new feature, new customer group, or expansion into another country can alter the system’s intended purpose or legal exposure. Legal teams, product owners, procurement, security, and operational staff should share responsibility for reviewing those changes.
Frequently Asked Questions
Does the EU AI Act apply to a US company?
It can. A US company may be covered when it places an AI system or model on the EU market, or when a system’s output is used in the EU, among other circumstances defined by the Act.
Is the EU AI Act stricter than US AI regulation?
It is more unified and explicitly risk-based across sectors. Whether it is stricter for a particular business depends on the system, the relevant US federal and state rules, and the obligations triggered in each jurisdiction.
Does the United States have a federal AI law?
The US has no single comprehensive federal statute governing AI across all private-sector uses. Existing federal laws and agency powers can apply, alongside state and local requirements.
What makes an AI system high-risk under the EU AI Act?
The Act defines high-risk categories, including specified uses in areas such as employment and education. Classification turns on the intended purpose and detailed legal criteria, so the technology alone does not decide it.
Can one compliance program cover both the EU and US?
A shared governance program can support both, but it should include jurisdiction-specific checks. EU conformity and documentation duties do not replace US obligations, and US controls do not automatically satisfy the AI Act.
Related
Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.
More cards in this drawer
- Puppets in TV Shows: Classic Proof They’re Better
- Best Streaming Remote: Top Picks for Every Setup
- Claude Pro Hidden Features Power Users Try
- Tesla Rental Prices: Why They’re So Low255.00 USD
- Best Family Western Movies to Stream Right Now
- Luxury V6 Sports Sedan: Can It Easily Top 100k Miles?