AI Regulation in the U.S.: States Push New Laws on Facial Recognition and Hiring Bias

AI regulation is heating up in the United States. With federal laws still limited, dozens of states are introducing their own AI governance bills. These efforts focus on two urgent issues: limiting facial recognition and preventing bias in hiring algorithms. As a result, state lawmakers are shaping the future of responsible AI use across the country.

Legislative Momentum: AI Regulation Expands Statewide

In 2025, nearly all U.S. states introduced some form of AI regulation. According to the NCSL, 38 states passed around 100 AI-related bills by midyear.

Key themes include:

• Ownership of AI-generated content – Arkansas clarified who owns AI-generated work when produced for employers.
• Infrastructure accountability – Montana’s “Right to Compute” law requires risk-management frameworks aligned with NIST’s AI standards.

These laws reflect growing concerns about transparency, fairness, and the unchecked spread of artificial intelligence.

Facial Recognition: Protecting Rights and Privacy

Facial recognition is one of the most heavily targeted areas of state AI regulation. By late 2024, 15 states had passed laws restricting its use, especially in government and law enforcement.

Moreover, many laws require companies to publish bias test results and ensure human review before making final decisions.

This wave of legislation stems from civil rights concerns. Studies consistently show facial recognition systems misidentify people of color at much higher rates, increasing risks of wrongful arrests and discrimination.

Hiring Bias: Independent Audits and Accountability

Hiring bias has also become a central focus of AI regulation.

• New York City’s Local Law 144, effective July 2023, requires annual third-party audits of automated hiring tools. Employers must also post transparency notices.
• Virginia is considering a ban on AI systems that cause unfair outcomes in hiring.
• Illinois already prohibits video-based AI screening under its AI Video Interview Act.
• Other states like Connecticut, Massachusetts, Washington, and Minnesota are proposing new transparency laws for hiring algorithms.

These efforts highlight a broad consensus: AI systems making employment decisions must be carefully monitored.

Texas Takes the Lead in AI Governance

Texas passed one of the most ambitious AI laws in 2025: the Texas Responsible AI Governance Act (TRAIGA).

It includes:

• Bans on AI-based social scoring and deepfake use with minors.
• Rules preventing AI systems from promoting crime or self-harm.
• Strict requirements for transparency and employee AI training.
• Penalties of up to $25,000 per violation.

Therefore, Texas is setting a precedent for other states by creating a comprehensive framework that covers both public and commercial AI use.

Federal Pushback vs. State Autonomy

At the federal level, Congress attempted to pause state-level AI regulation. A House bill even proposed a 10-year moratorium that would block states from passing AI-related laws.

However, this move faced strong criticism. Attorneys general, civil rights groups, and state leaders argued that local governments must retain power to address AI risks. As a result, the Senate rejected the moratorium in mid-2025, allowing states to continue developing their own laws.

Spotlight on Key States

• Colorado – Reviewing its groundbreaking 2024 healthcare AI law to balance costs and safeguards.
• California – Advancing bills on chatbot safety, automated decision rights, and public AI resources (CalCompute).
• New York – Mandating agency AI disclosures and audits, though oversight remains inconsistent.
• Montana – Recently banned manipulative AI uses and required human sign-off on AI-driven government actions.

Why State-Level AI Regulation Matters

State-level AI governance shows that legislators are serious about protecting civil rights, ensuring fairness, and building public trust.

Yet, the patchwork of different state laws also creates challenges for businesses, which must adapt to varying rules. The tension between federal uniformity and state independence will continue shaping the future of AI regulation in the U.S.

Ultimately, effective oversight is essential to ensure that artificial intelligence serves society responsibly, transparently, and fairly.