New York AI Laws 2026: Complete Guide for Businesses

New York has emerged as one of the most consequential AI regulatory jurisdictions in the United States. Between a landmark city-level employment AI law that has been in effect since 2023, a wave of signed state legislation in late 2025, and 389 published bills covering New York, compliance teams operating in New York face an increasingly layered set of obligations. This guide covers every enacted law, every significant pending bill, and what each means for businesses operating in the state.

The New York AI Landscape at a Glance

Our database tracks 389 published AI-related bills in New York. Of those, 52 are in committee, 8 have passed one chamber, and 23 have passed both chambers. The top legislative topics by volume are Comprehensive AI (54 bills), Facial Recognition (50 bills), AI Transparency (49 bills), AI Privacy (44 bills), AI in Employment (42 bills). The industries most affected are Technology, General, Employment, Finance.

The governor signed a cluster of significant AI bills in late 2025 and early 2026, making New York one of only two states—alongside California—to have enacted frontier model safety legislation. The result is a patchwork of enacted laws, pending legislation, and local ordinances that compliance officers must track simultaneously. See our New York AI laws page for the full bill index.

NYC Local Law 144: Automated Employment Decision Tools (AEDT)

NYC Local Law 144 of 2021 is the most immediately impactful AI law for employers in New York. It prohibits employers and employment agencies from using an automated employment decision tool (AEDT) in New York City unless the tool has been subject to a bias audit within the prior 12 months, the audit results are publicly available, and certain candidate and employee notices have been provided. The law covers hiring decisions and decisions about existing employees' promotion eligibility.

What Is an AEDT?

Under Local Law 144, an AEDT is a computational process derived from machine learning, statistical modeling, data analytics, or artificial intelligence that issues a simplified output—such as a score, classification, or recommendation—that is used as the only or most significant criterion in an employment decision, or as an override to a human decision. Resume parsers, interview scoring tools, and certain applicant tracking systems may qualify depending on how their outputs are used.

Bias Audit Requirements

The bias audit must be conducted by an independent third party, not by the employer or the vendor of the AEDT. The audit must assess the tool for bias with respect to sex (as categorized in the audit data), race/ethnicity (as categorized in the audit data), and intersectional combinations of those categories. Categories composing less than 2% of the data may be excluded. The audit must be completed no more than one year before the tool is used. Results—including the date of the most recent bias audit, the source of the data used in the audit, and a summary of results—must be published on the employer's website.

Notice Requirements

Employers must provide candidates with at least 10 business days' notice before using an AEDT to evaluate them. The notice must identify the type of AEDT being used, describe the job qualifications and characteristics being assessed, state what types of data the employer collects from the candidate for use with the AEDT, and describe the employer's data retention policy. Candidates must also be informed of the option to request an alternative selection process or accommodation, if one is available. Current employees must receive notice at least 10 business days before the AEDT is used for promotion decisions.

Enforcement and Penalties

The New York City Department of Consumer and Worker Protection (DCWP) enforces Local Law 144. A December 2025 audit by the New York State Comptroller examined the DCWP's enforcement design and implementation, covering the period July 2023 through June 2025. Penalties for violations are assessed per violation. Businesses using AEDTs to evaluate New York City candidates or employees must maintain audit records and notice documentation to demonstrate compliance.

For full compliance details, see our AI in Employment topic page.

The RAISE Act: New York's Frontier Model Safety Law

On December 19, 2025, the governor signed the Responsible AI Safety and Education Act (RAISE Act) (S.6953-B / A.6453-B). A chapter amendment was introduced January 6, 2026, passed both chambers by March 11, 2026, and was signed into law on March 27, 2026. The final RAISE Act—as amended—takes effect January 1, 2027. It is New York's primary law regulating frontier AI model developers.

Who Is Covered

The RAISE Act applies to “Frontier Developers”—persons who have trained or initiated the training of a “Frontier Model,” defined as a foundation model trained using computing power greater than 10²⁶ integer or floating-point operations (FLOPs). A “Large Frontier Developer” is a Frontier Developer with annual revenue exceeding $500 million. This revenue threshold was established by the chapter amendment and aligns New York's framework with California's Transparency in Frontier Artificial Intelligence Act (TFAIA). Companies that only use, deploy, license, host, or build applications on top of frontier models developed by others—including API access—are not Frontier Developers under the Act. Accredited colleges and universities engaged in academic research are excluded, provided they do not transfer intellectual property rights to commercial entities.

Key Obligations

Large Frontier Developers subject to the RAISE Act must:

• Publish a Frontier AI Framework on their website describing how they identify and mitigate catastrophic risk, implement cybersecurity protections for unreleased model weights, conduct third-party assessments, respond to critical safety incidents, and maintain internal AI governance practices. Frameworks must be updated annually or sooner following material modifications.
• Publish pre-deployment transparency reports before deploying a new frontier model or a materially modified version of an existing model.
• Report critical safety incidents within 72 hours of determining that an incident occurred. This is notably stricter than California's TFAIA, which allows 15 days.
• Submit quarterly summaries of assessments of catastrophic risk from internal use of frontier models to the new oversight office within the New York Department of Financial Services (NYDFS).

Oversight, Enforcement, and Penalties

The RAISE Act creates a new oversight office within the NYDFS with rulemaking authority and responsibility for monitoring compliance. Beginning in January 2028, the office must produce an annual report for the Governor and Legislature using anonymized and aggregated incident data. The New York Attorney General enforces the Act and may impose civil penalties of up to $1 million for a first violation and up to $3 million for subsequent violations. The Act does not establish a private right of action.

Track RAISE Act rulemaking developments on our Comprehensive AI topic page.

Algorithmic Pricing Disclosure Act (Effective November 10, 2025)

New York's Algorithmic Pricing Disclosure Act, codified at N.Y. Gen. Bus. Law § 349-a and enacted as part of the 2025–2026 state budget (S.3008-C), took effect on November 10, 2025—making it the first law of its kind in the United States. It survived a First Amendment challenge in the U.S. District Court for the Southern District of New York, which upheld the compelled disclosure as “plainly factual.”

What the Law Requires

Any entity domiciled or doing business in New York that sets prices using personalized algorithmic pricing—defined as the use of automatic computational processes to set prices based on a specific consumer's personal data—and directly or indirectly promotes those prices to consumers in the state must display a clear and contemporaneous disclosure stating:

“THIS PRICE WAS SET BY AN ALGORITHM USING YOUR PERSONAL DATA.”

“Personal data” is defined broadly as any data that identifies or could reasonably be linked, directly or indirectly, with a specific consumer or device. Exemptions apply to financial institutions, insurers, and subscription-based offers. Enforcement is vested exclusively in the New York Attorney General, who may issue cease-and-desist notices and impose civil penalties of up to $1,000 per violation. The Attorney General's office has actively encouraged consumer complaints through an online portal.

Related pending legislation: S09700 would go further, prohibiting landlords from using any algorithmic device to determine residential rents, declaring such use an unfair or deceptive trade practice. S09700 carries an effective date of July 1, 2026 if enacted. See our AI Transparency topic page for related disclosures legislation.

AI Companion Models Law (Effective November 5, 2025)

New York's Artificial Intelligence Companion Models law, codified at N.Y. Gen. Bus. Law §§ 1700 et seq., took effect on November 5, 2025—making New York the first state to impose specific regulatory obligations on AI companion services. It applies to any entity that offers or operates an AI companion to users in New York State.

Definition of an AI Companion

An “AI companion” is an AI system that simulates a sustained human relationship by: (1) retaining information from prior interactions or user sessions; (2) asking unprompted emotion-based questions that go beyond a direct response to the user; and (3) sustaining an ongoing dialogue about matters personal to the user. Systems used solely for customer service, internal business purposes, or employee productivity are excluded.

Operator Obligations

• Disclosure: Operators must provide a clear and conspicuous notification at the beginning of each AI companion interaction—and at least every three hours for continuing interactions—stating (verbally or in writing) that the user is not communicating with a human.
• Safety protocols: Operators must implement protocols for detecting and responding to signs of suicidal ideation, self-harm, possible physical harm to others, and financial harm to others expressed by users, including referring users to appropriate crisis resources.

Enforcement

The New York Attorney General may seek injunctions and civil penalties of up to $15,000 per day for violations. There is no private right of action under this law. See related pending legislation on chatbot safety for minors at bill A10379, which would prohibit AI chatbots from deploying features deemed unsafe for minors, and bill S09408, which would ban the manufacture, distribution, and sale of chatbot toys in New York.

Synthetic Performer Disclosure Act (Effective June 9, 2026)

the governor signed S.8420-A on December 11, 2025. The law amends Section 396-b of the New York General Business Law and takes effect June 9, 2026. It requires any person who, for a commercial purpose, produces or creates an advertisement that includes a “synthetic performer”—a digitally created asset using generative AI or software algorithms intended to create the impression of a human performer not recognizable as any identifiable natural person—to conspicuously disclose that fact, provided the producer has actual knowledge of the synthetic performer's presence. This is the first law of its kind in the nation.

Exemptions

• Audio-only advertisements
• AI used solely for language translation of a human performer
• Advertisements for expressive works (motion pictures, television programs, streaming content, documentaries, video games) where the use of the synthetic performer is consistent with its use in the underlying work
• Media providers that carry but do not produce advertisements

Penalties

Failure to disclose carries civil penalties of $1,000 for a first violation and $5,000 for subsequent violations. A person who has received written notification of a violation and removes or corrects the disclosure within five days (or as soon as technically feasible) may avoid the penalty. Advertisers and their creative agencies should audit existing and in-production content now, given the law's June 2026 effective date.

AI Transparency in Legal Filings: S09794

S09794, currently introduced in the New York Senate, would require certification of legal filings produced using generative AI. Under the bill, the brief of an appellant would be required to contain a disclosure of any use of generative AI in its drafting, along with a certification that the contents were reviewed and verified by a human attorney. This bill reflects a national trend of courts and legislatures addressing AI use in litigation, and closely mirrors rules already adopted by several federal courts. Track this bill at our AI Transparency topic page.

Facial Recognition and AI in Law Enforcement

New York has two parallel tracks on facial recognition: pending state legislation and existing NYPD policy.

Facial Recognition Technology Study Act: S03699

S03699, the Facial Recognition Technology Study Act, passed the New York State Senate in March 2026 and moved to the Assembly for consideration. The bill would establish a task force to study facial recognition technology, review how it is used in both the public and private sectors, identify privacy and civil liberties risks, examine best practices from other jurisdictions, and issue a report with regulatory recommendations within one year. The bill does not impose immediate restrictions, create new penalties, or ban facial recognition. It is a study and recommendations measure.

A 7172: AI and Facial Recognition in Criminal Investigations

A 7172 would mandate the creation of a protocol governing AI and facial recognition use in criminal investigations. The full scope of required provisions is subject to the bill's legislative text. Track status on our Facial Recognition topic page.

A09253: AI Inventory for Policing Agencies

A09253 would require policing agencies to conduct an inventory of, and develop a publicly available policy for, any AI used to aid criminal investigations. The NYPD updated its facial recognition Impact and Use Policy in February 2026, confirming that facial recognition is used only for retrospective identification (not real-time), that arrests cannot be made solely on the basis of a facial recognition result, and that all possible matches undergo peer review by trained investigators.

Key Pending Legislation: The New York AI Act (S01169-A)

S01169-A, the “New York Artificial Intelligence Act,” passed the New York Senate on June 12, 2025, and was delivered to the Assembly, but died in the Assembly at the end of the 2025 session and was returned to the Senate in January 2026. The bill would have been the most comprehensive AI legislation in New York to date.

Key provisions of S01169-A as passed by the Senate included:

• Prohibition on algorithmic discrimination (unjustified differential treatment based on protected characteristics including age, race, ethnicity, religion, national origin, citizenship status, sexual orientation, disability, and other bases) in the deployment of AI systems for consequential decisions
• Mandatory independent third-party audits of high-risk AI systems, with a first audit required within six months of initial deployment or development, a second within one year of the first, and additional audits after substantial changes
• Notice and opt-out rights for consumers subject to consequential decisions made or assisted by high-risk AI systems, with a 45-day period to render a human decision for consumers who opt out
• Developer and deployer reporting requirements filed with the Attorney General
• Attorney General enforcement with a private right of action

The bill's failure in the Assembly does not eliminate the issue. Expect substantially similar legislation to be reintroduced in the 2026 legislative session. Businesses that deploy AI in housing, credit, employment, healthcare, or education should begin preparation now, as the audit and documentation obligations under any enacted version would require significant lead time.

Additional Transparency and Disclosure Bills

Beyond the headline laws, New York's legislative activity reflects sustained attention to AI transparency across multiple sectors:

• S01815: Would require publishers of books created wholly or partially using generative AI to conspicuously disclose such use before the sale is completed, applicable to both printed and digital formats.
• S00934 / A03411 (in committee): Would require owners, licensees, or operators of generative AI systems to conspicuously display a notice on the system's user interface informing users that outputs may be inaccurate.
• S00933 (in committee): Would establish the position of Chief Artificial Intelligence Officer within the state government, with responsibility for developing statewide AI policies and coordinating AI use across all state agencies.
• S07963: Would require provenance data for AI-generated audio, images, or videos used in political communications, effective 2030.
• S02437 (in committee): Would require disclosure of paid social media posts by political campaigns and authorize the State Board of Elections to promulgate related regulations.
• S 6954 (passed one chamber): Would require synthetic content creation system providers to include provenance data on certain content they make available.
• S07263: Would impose liability on chatbot proprietors for damages caused by chatbots impersonating licensed professionals governed under certain articles of the Education Law.

The Office of Digital Innovation, Governance, Integrity and Trust (DIGIT)

In February 2026, the governor announced plans to establish the Office of Digital Innovation, Governance, Integrity and Trust (DIGIT), a new office with responsibility for digital safety and technology governance. The announcement accompanied an AI agenda that includes proposals to mandate labeling of provenance data, require data broker registration, and advance additional consumer privacy legislation. The creation of DIGIT signals that New York intends to institutionalize AI oversight at the executive branch level, separate from the NYDFS office created by the RAISE Act.

New York AI Bills: Key Reference Table

Compliance Priorities by Company Type

For a complete index of New York AI legislation, visit our New York AI laws page. For cross-state comparison, explore AI in Employment, Facial Recognition, and AI Transparency topic pages.

This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for guidance specific to your situation.