Summary#
Int. 1116-2026, sponsored by Council Member Virginia Maloney, would amend Title 10 of New York City’s Administrative Code by adding a private right to sue over certain harms caused by third-party misuse of AI models. Its stated aim is to let people seek remedies when harm was foreseeable and the provider did not use reasonable safeguards. The bill has been referred to the Committee of the Whole and is not law.
- A person could sue an AI provider for harm occurring in New York City if a third party’s misuse or malicious use of the provider’s model caused the harm, the provider failed to use reasonable safeguards against that misuse, and the harm was foreseeable.
- The definition of AI provider covers any person who makes an AI model available for commercial or public use.
- A person who brings a claim could seek damages, including punitive damages, an injunction, and other appropriate remedies.
- The bill would take effect immediately if enacted. It does not create a new agency or set out a city-run claims process.
What it means for you#
- People harmed in New York City: You could bring a civil claim against a provider if all three conditions in the bill are met. The bill does not guarantee that a claim would succeed.
- AI providers: You could face a lawsuit over harm caused by a third party’s misuse of a model you made available. The bill does not set a specific safeguard checklist.
- Third-party users: The bill does not create a claim against the user; it creates a claim against the provider under the stated conditions.
Money#
No cost information is in the available material.
- The bill authorizes people to seek damages, including punitive damages, but gives no amounts or estimate of likely awards.
- No public spending, staffing, fees, fines, or compliance-cost estimates are provided.
What is unclear#
- The bill does not define “misuse,” “malicious use,” “reasonable safeguards,” or “foreseeable.”
- It does not explain how a court should assess whether safeguards were reasonable or harm was foreseeable.
- The text does not set limits on who may bring a claim beyond “a person,” or specify deadlines or procedures.
- The available material does not show how this claim would interact with existing legal claims or rules.
Case for#
- The Council’s summary describes a route to seek a remedy when a provider could foresee harm from third-party misuse but failed to use reasonable safeguards.
- A possible argument for the bill is that providers could be held financially accountable for certain foreseeable harms linked to misuse of models they make available.
Case against#
- One concern is that key terms, including “reasonable safeguards” and “misuse,” are not defined, leaving important questions for courts to resolve.
- The bill allows claims for “any harm” meeting its conditions and does not specify damages limits; the practical scope of liability is unclear.