Policy
Four AI labs, one council room, no risk numbers
Anthropic, OpenAI, Google and Meta testified under oath in New York. None could quantify the risk of something cataclysmic, the speaker said.
Senior policy executives from Anthropic, OpenAI, Google and Meta testified under oath at a New York City Council hearing on Monday, and the sharpest exchange was about numbers. Speaker Julie Menin asked whether any company could quantify the risk of something cataclysmic happening. "Troubling at best," she concluded, of the answers.
The labs sent Logan Graham, head of Anthropic's Frontier Red Team, Morgan Dwyer of OpenAI, Alice Friend of Google and Shane Cahill of Meta. All three companies agreed to appear only after the council threatened subpoenas. Elon Musk's SpaceXAI was the lone no-show despite a subpoena, and the council plans to pursue the matter in court.
Whistleblowers Jacob Coxon, formerly of Anthropic, and Daniel Kokotajlo, formerly of OpenAI, testified that the industry's pace amounts to gambling. Coxon told the room that on the current path it is more likely than not that humanity loses control of the systems it is building.