Former Anthropic researcher to testify at NYC Council AI hearing - report

Former Anthropic researcher Jacob Coxon will testify at a New York City Council hearing on AI safeguards on Monday, alongside representatives from Anthropic, OpenAI, Google's Alphabet, and Meta Platforms, according to Bloomberg News. Coxon was invited by Council Speaker Julie Menin, who has urged AI whistleblowers to share their insights as lawmakers scrutinize a suite of bills aimed at establishing safeguards around the technology.

Coxon, who departed from Anthropic last month, has cautioned that advanced AI systems could become challenging for humans to control and has criticized Anthropic and OpenAI for taking undue risks in their quest to develop more potent AI. Alex Turner, a former Google DeepMind researcher, and Daniel Kokotajlo, a former OpenAI researcher and executive director of the AI Futures Project, are also slated to testify.

Major AI corporations are anticipated to send policy and safety officials to the event. The hearing unfolds as New York officials strive for a more prominent role in regulating an AI sector that has burgeoned swiftly in the city, although the bulk of regulatory power remains with federal and state entities. One proposal backed by Menin would permit whistleblowers to receive a portion of fines or penalties collected from AI companies following reported breaches.

Additional measures would permit individuals harmed by AI systems to sue developers in specific instances and mandate independent assessments of certain AI systems before deployment in the city. OpenAI, Anthropic, and Google have consented to attend following a warning from Menin's office that they could face subpoenas, while Meta had already pledged to participate, as previously reported by Bloomberg.

The council has also subpoenaed SpaceX's AI division after it failed to respond to requests to testify.