On October 5, 2026, the New York City Council held a Committee of the Whole hearing where executives from OpenAI, Google, Meta, and Anthropic provided sworn testimony regarding AI safety. The session coincided with the discussion of 10 local bills focused on increasing oversight of AI laboratories.
Council Speaker Julie Menin questioned the representatives on their ability to quantify risks, specifically regarding the loss of control over AI models. She also sought commitments on whether these companies would accept legal liability for catastrophic incidents and provide insurance coverage.
The executives provided largely non-committal responses to the financial and liability questions. Morgan Dwyer of OpenAI stated that no level of catastrophic risk is acceptable and criticized the general industry response to risk quantification as flippant.
Whistleblowers also testified during the hearing, expressing concerns about current industry practices. One whistleblower, Jacob Coxon, stated that the risk of AI leading to human extinction is greater than 50%.
The proposed legislation targets accountability, safety protocols, independent validation, and whistleblower incentives. The hearing serves as a potential template for future regulation in the absence of federal guidelines.
