New York City Council Proposes AI Kill Switches and Fines

The New York City Council introduced a legislative package on September 25 to regulate artificial intelligence companies operating within the city. The proposed bills would mandate the installation of kill switches—human overrides capable of shutting down systems—and require third-party validation for bias, privacy, and security before AI tools can be marketed or deployed in the five boroughs. Non-compliance would trigger fines of $25,000 per offense.

Council Speaker Julie Menin unveiled additional measures including a first-of-its-kind program to provide cash rewards to whistleblowers using funds collected from fines. The legislation would also allow citizens to sue AI companies for foreseeable harm caused by jailbroken tools and require city contractors to report safety incidents within 24 hours. Menin stated that the city intends to remain the AI capital of the world while establishing guardrails to build public trust, arguing that the industry should not self-regulate.

Menin has requested that leaders from OpenAI, Google, Meta, Anthropic, and SpaceX attend a Committee of the Whole hearing on October 5 to discuss technology risks, though sources indicate none are likely to attend. This municipal effort follows a recent mandate from Governor Kathy Hochul requiring AI companies to register in a state portal and report safety incidents within 72 hours, reflecting a broader push for local oversight amid a lack of federal regulation.