World's most comprehensive binding AI regulation. Risk-based framework: bans unacceptable-risk AI (social scoring, real-time biometric surveillance); imposes strict requirements on high-risk AI in employment, credit, education, and essential services including bias testing, technical documentation, human oversight, and fundamental rights impact assessments.

Gives individuals the right not to be subject to decisions based solely on automated processing that produce legal or similarly significant effects. Controllers must implement suitable safeguards and provide the right to human review.

Requires employers to publish gender pay gap data and conduct joint pay assessments where a gender pay gap of 5% or more exists and cannot be justified. AI-based compensation tools that produce unjustified gaps are directly implicated.

Prohibits employment discrimination on the basis of race, color, religion, sex, or national origin. Under Griggs v. Duke Power, facially neutral AI tools that produce a disparate impact on a protected class trigger scrutiny — employers bear the burden of demonstrating business necessity.

Technical assistance documents establishing that AI hiring tools violating the ADA (screening out individuals with disabilities) or producing disparate impact are actionable. Vendors and employers share liability. The iTutorGroup case ($365K settlement) was the first EEOC action targeting an AI hiring tool.

Directs federal agencies to develop AI safety standards and address AI bias. Requires mandatory reporting by frontier model developers and development of AI safety standards by NIST. Signals direction of federal rulemaking.

The FTC has authority to prohibit unfair or deceptive practices involving AI under Section 5 of the FTC Act. Discriminatory AI in credit scoring, insurance pricing, and ad targeting causing consumer harm is actionable. The CFPB requires adverse action notices to explain AI-based credit decisions.

Requires employers using AEDTs in NYC to conduct independent bias audits annually, publish impact ratios by sex/race/ethnicity on their website, and notify candidates that an AEDT will be used. The most detailed AI hiring regulation in force in the US.

Requires employers using AI to analyze video interviews to notify candidates, explain how the AI works, obtain consent, and delete videos within 14 days of request. One of the first state laws regulating AI in hiring.

Prohibits insurers from using algorithms that unfairly discriminate based on protected characteristics. The Colorado Division of Insurance can audit models and require corrective action. First US state law explicitly targeting algorithmic discrimination in insurance.

Most comprehensive US state AI accountability law. Requires annual impact assessments, individual notification, right to opt out of solely automated decisions, and appeal mechanism. Enforced by the Attorney General with a private right of action.

First US state law restricting facial recognition in employment screening. Employers must obtain written consent before using facial recognition technology during interviews.