What are the odds that artificial intelligence leads to a disaster? New York City lawmakers asked four of the world’s biggest AI companies that question this week. Not one of them gave a number.
The exchange came during a marathon City Council hearing on AI safety on Monday (Oct. 5), Route Fifty reported. Google, Meta, Anthropic and OpenAI sent representatives to testify. Their testimony lasted several hours. The hearing stretched across 14 panels and was still going when Route Fifty published its story.
One major player was missing. Elon Musk’s SpaceXAI ignored a subpoena to appear. The council is now taking legal action to force the company to comply.
Council Speaker Julie Menin led much of the questioning. Menin is a former regulatory lawyer, and she pressed the company representatives the way an attorney might question the other side in court.
Menin asked each company to estimate the risk artificial intelligence poses in a worst-case scenario. None of them offered an answer. She then asked whether the companies would release AI systems that fail their own safety tests or tests run by outside groups. The representatives talked about their internal reviews and their commitment to safety. But they did not give a yes or a no. Menin said she would treat their answers as dodges.
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
According to Route Fifty, Menin said she learned more from what the companies left unsaid than from anything they told the council. Several other council members said they heard nothing new at all.
“I think it’s disappointing that they can’t seem to answer some of the most basic, rudimentary questions,” Menin told City & State around the fourth hour of the hearing.
The mood in the room shifted sharply depending on who held the microphone. The first panel of the day featured former researchers from leading AI companies who have since become whistleblowers. Their message was stark. They warned that a catastrophic event is a real possibility. They also said the companies are racing ahead without enough safeguards in place.
Jacob Coxon, a former Anthropic researcher, drew international attention last month with his warnings. At the hearing, he said AI could lead to major gains in science and improve people’s lives. But on the current path, he said, he believes it is more likely than not that humans lose control of the technology. He warned that the result could be human extinction.
The company representatives struck a calmer tone. Most of them work on policy rather than building artificial intelligence systems. That may explain why they steered clear of hard numbers.
So, what comes next? The hearing gave the council a chance to show it is acting on AI while Washington stays on the sidelines. Route Fifty reported that the council wants to place itself at the front of AI regulation. That push comes as federal action remains out of reach.
The council’s legal fight with SpaceXAI is the next test. It will show how much power a city has to make a major tech company answer questions in public. The outcome could shape how other cities and states approach AI companies that refuse to cooperate.
More testimony was still to come when Route Fifty went to press. For the AI industry, the takeaway from Monday was simple. Local lawmakers want straight answers about risk, and they plan to keep asking until they get them.
Add as Preferred Source
