New York City Council holds AI safety hearing; former employees and representatives of four labs testify

The New York City Council held a hearing on AI safety, where former employees of Anthropic, OpenAI, and Google DeepMind testified, saying the companies have a reckless culture and are racing to develop technology they cannot control. Former Anthropic engineer Jacob Coxon said that on the current path, humans are more likely to lose control of AI, possibly even ending in human extinction. Current employees of OpenAI, Anthropic, Google, and Meta also testified, saying the systems are already improving daily life and emphasizing their focus on safety. City Council Speaker Julie Menin asked company representatives to quantify the risk of worst-case disaster scenarios; OpenAI's Morgan Dwyer said she did not know a specific probability and stated that 1%, 10%, or 20% would all be unacceptable, and Menin criticized the answer as reckless. The City Council is considering multiple AI regulation proposals, including requiring companies to disclose information about AI tools, introducing third-party approval for municipal AI systems, and allowing whistleblowers to share violation fines.