Home / Technology / AI Hearing: NYC Council Gets No Answers on Catastrophe Risks
AI Hearing: NYC Council Gets No Answers on Catastrophe Risks
6 Oct
Summary
- Major AI firms couldn't quantify catastrophic risks or provide safety assurances.
- A subpoenaed SpaceX AI representative failed to attend the NYC hearing.
- Whistleblowers warn of AI's existential threat and companies' recklessness.
New York City Council members held a hearing to question representatives from leading AI companies, including OpenAI, Anthropic, Meta, and Google, regarding the potential catastrophic risks posed by artificial intelligence.
During the hearing, companies were unable to quantify the likelihood of disastrous AI events or detail their insurance coverage for such scenarios. When pressed by Council Speaker Julie Menin, OpenAI's representative admitted to not knowing the exact risk percentage of a catastrophic AI failure, stating that any chance was unacceptable.
The session also revealed that a representative from Elon Musk's SpaceX AI arm, despite a subpoena, did not attend. Council Speaker Menin indicated she would seek judicial intervention.
Testimony from former AI employees, such as Jacob Coxon from Anthropic, presented a stark warning. Coxon described the companies' culture as "extremely reckless," likening their "move fast, break things" approach to that of a photo-sharing app, which he argued is dangerously inappropriate for developing powerful AI.
Despite the concerns, AI companies highlighted the technology's benefits, including advancements in medicine, cybersecurity, and economic growth. However, proponents like Mr. Coxon also acknowledged the potential for AI to lead to human extinction if not controlled.
New York City Council has proposed legislation, including requirements for independent validation and human shut-down capabilities for AI models sold in the city. Mayor Zohran Mamdani expressed support for stronger federal safeguards and has implemented a moratorium on generative AI for young students.
