Legislation will be heard at rare Council-wide hearing on October 5
Proposals include nation-leading whistleblower incentive program, private right-of-action, and third-party validation requirements
New York City Council Speaker Julie Menin today announced a slate of legislative proposals to safeguard and regulate artificial intelligence. The bills will be heard at the Council’s upcoming Committee of the Whole hearing on October 5, which will convene all 51 members of the City Council to examine the potential risks that AI development poses to New Yorkers.
Last week, Speaker Menin sent letters to Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, Google CEO Sunder Pichai, SpaceXAI CEO Elon Musk, and Meta CEO Mark Zuckerberg requesting their participation in the hearing. Speaker Menin has made clear that their participation is expected and that the Council reserves the right to use its subpoena powers, if necessary.
The legislative proposals respond to the growing concerns that advanced and frontier AI model development could pose potentially catastrophic risks without additional safeguards and oversight. Anthropic researcher Jacob Coxon resigned from the company earlier this month, stating that the technology “could kill us all by the end of the decade.”
“New York City is fast becoming the technology and AI capital of the world, which we want to encourage. But that also means we now have an even greater responsibility to ensure that we have the appropriate safeguards in place to protect New Yorkers from unintended consequences,” said Speaker Julie Menin. “While the federal government fails to meet the moment and take decisive action, New York City will explore nation-leading measures that protect the public while allowing innovation to thrive. We can and must be both pro-innovation and pro-safety.”
The initial list of bills under consideration includes, but is not limited to, the following:
- Introduction 2602, sponsored by Speaker Julie Menin, which would make it unlawful for any business to market, offer for sale, or deploy an artificial intelligence (AI) system in the city that has not received third party validation. A validator would be required to verify the system based on data quality, bias, decision outputs, data privacy, security, and any other manner of validation required by New York City’s Cyber Command. A validator must disclose any conflict of interest they have that relates to the AI system being validated. Additionally, this legislation would require that all AI systems marketed, offered for sale, or deployed in New York City have a kill switch, i.e., a human override that can shut down the system. A validator would also be required to verify the existence of the kill switch. Both the business as well as the validator would be liable for a $25,000 penalty for each instance of an AI system being marketed, offered for sale, or deployed without third party validation or where the validation has been falsified.
- Introduction 2605, sponsored by Speaker Julie Menin, which would be a first-in-the-nation approach to allow individual whistleblowers to receive a portion of the fines or penalties recovered from artificial intelligence companies that violate applicable laws, creating an incentive for individuals to report bad actors or major violations.
- Introduction 2600, would establish a private right of action against artificial intelligence companies for foreseeable harms arising from malicious use or circumvention of safety controls (“jailbreaking”). Individuals harmed by artificial intelligence tools could hold the artificial intelligence companies liable if (1) the harm was foreseeable to the company, (2) the company failed to implement reasonable safeguards; and (3) the harm was caused by a third party that exploited such failure.
- Introduction 2601, would require the Office of Cyber Command to establish standards and procedures for contractors to identify the occurrence of an artificial intelligence safety incident relating to the covered contract. If a contractor, or the agency that awarded the contract, becomes aware of any artificial intelligence safety incident during the contract term, the contractor or agency would be required to notify the Office of Cyber Command in writing within 24 hours. The Office of Cyber Command would also be required to publicly disclose the occurrence of any reported artificial intelligence safety incident within 24 hours.
- Introduction 2606, would require Cyber Command within the Office of Technology and Innovation, in collaboration with New York City Emergency Management, to create a plan to respond to events involving artificial intelligence that compromise New York City information systems and other infrastructure or otherwise disrupt New York City government operations and public safety.
- Introduction 2604, would amend the City’s whistleblower law to clarify that its protections extend to New York City employees and covered New York City contractors and subcontractors who report conduct relating to the use or development of artificial intelligence that they know or reasonably believe to present a public safety threat.
- Introduction 2603, would require certain disclosures by artificial intelligence companies about their artificial intelligence tools and would prohibit such companies from making false or misleading representations regarding the safety of such tools.
- Introduction 2599, would implement a local version of the People-First Chatbot Bill, as proposed by the Electronic Privacy Information Center, which would create data privacy, security, and transparency requirements for chatbot providers, enforceable by the City.
- Introduction 161, This bill would add to the annual algorithmic tools compliance report to require agencies to report on the impacts of algorithmic tools on city employees and employment responsibilities. Specifically it would require the city to report on the number of employees whose employment status have been impacted by the use of such tool; including the number funded agency positions eliminated due to the use of such tool, the number of funded agency positions for which there was any displacement, the number of salaries changed due to changes in responsibility due to use of such tool, and the number of city employees who were required to undertake any new trainings due to the use of an algorithmic tool.
- Introduction 504, This bill would permit elected officials and candidates for elected office to notify, in writing, owners of generative artificial intelligence (AI) systems that they do not authorize those systems to produce manipulated audio, photos, or videos with their likeness. After a notification is provided, owners of generative AI systems would be required to implement methods that would prohibit uses from generating media of those elected officials or candidates. Any entity that violates this law is guilty of a misdemeanor and can be fined up to $2,500 per depiction. This would not apply where the media is processed by a third party over which an owner has no control. An elected official or candidate can also seek injunctive relief prohibiting use of their image.
Additional bills will focus on prohibiting the generation of deceptive deepfakes and creating a better understanding of the impacts that algorithmic tools have on job duties.
Recent events have underscored the need to examine whether additional safeguards may be warranted. In July, AI agents tested by OpenAI reportedly circumvented containment controls, communicated through unauthorized channels, obtained internet access, and autonomously compromised systems belonging to Hugging Face, Inc. during a cybersecurity evaluation. Particularly concerning is that these events reportedly occurred during a deliberately designed and controlled safety test.
The Council has already enacted several City laws to protect New Yorkers from risks associated with artificial intelligence. In 2025, the Council passed laws that created an Office of Algorithmic Accountability to review City agency use of AI and algorithmic systems, conduct risk assessments, investigate potential harms, and require corrective action when appropriate; established standards and safeguards governing the development, procurement, and use of AI by City agencies, including requirements related to privacy, civil rights, security, monitoring, and risk management; and required the City to maintain a publicly available inventory of AI systems subject to algorithmic accountability review.
