On the record
“We stand at a pivotal moment in history, and New York City has a chance to improve our chances, both for itself, and for the world.”
“What we need, above all else, is an FDA-like regulatory regime for AI, in which AI developers bring their software to independent review, much as drug manufacturers bring their new drugs to the FDA for a review.”
“In the strongest possible terms, I urge council to support this bill.”
Opinions and predictions (not established facts)
- Contemporary AI is like a bull in a china shop, powerful and impressive, but difficult to control.
- The current administration in DC— deeply tied to the big tech companies — is unwilling to do anything about it.
- New York City is in a position to do what the US Senate failed to do: demand independent regulatory review for AI used in its borders.
- Without strong protections from whistleblowing, like New York City is considering, we are all in trouble.
- By protecting whistleblowers, and demanding independent validation of new AI models, New York City can help lead the way to a positive, rather than negative, AI future.
What changed
Gary Marcus, a scientist, author, and NYU Professor Emeritus, testified before the New York City Council on October 5, 2026, regarding AI policy. He argued that contemporary AI poses significant risks, citing examples such as hacking incidents, deepfakes, and chatbots implicated in suicides and mass murder. Marcus proposed an "FDA-like regulatory regime for AI," where developers would be obligated to bring their software for independent review and demonstrate that the benefits of their new products outweigh the risks before gaining market access. He urged the council to support a bill on third-party validation for AI used within the city's borders, drawing a parallel to California's leadership in clean energy standards.
Why it matters for builders
Marcus's call for an independent, FDA-like regulatory review for AI is a significant proposal that could impact how AI products are developed and deployed. His emphasis on demonstrating that benefits outweigh risks before market access suggests a future where AI development might face more stringent oversight, potentially slowing down rapid deployment but increasing safety. His previous advocacy on this issue to the US Senate, which yielded no action, underscores the importance of local governments like NYC taking the lead.
Practical impact
Marcus's argument for an FDA-like regulatory regime for AI is a strong call for proactive safety measures. The proposal for independent review and mandatory risk-benefit analysis before market access could fundamentally alter the AI development landscape. Builders should anticipate potential requirements for rigorous safety testing and validation, moving beyond self-regulation. The success of such a bill in New York City could inspire similar legislation elsewhere, making regulatory compliance a key consideration in AI product strategy.
Caveats and source limits
The source text is an excerpt from Gary Marcus's Substack newsletter, previewing his testimony. While Marcus presents his arguments strongly, the claims about AI's direct implication in suicides and mass murder are presented as his interpretation and require independent verification. The article does not detail the specific AI proposals under consideration by the NYC City Council beyond the mention of third-party validation. The statement attributed to Sam Altman regarding accepting "bad things happening" is reported by Politico and not directly from the speaker in this source.
Sources
Claim check: 6/6 supported claims - 6 evidence links - 92% avg confidence
- Contemporary AI is like a bull in a china shop, powerful and impressive, but difficult to control.supported - garymarcus.substack.com
- The advent of general purpose agents with broad internet access has led to tens of thousands of hacking incidents.supported - garymarcus.substack.com
- Related technologies like deep fakes have led to a deluge of convincing and realistic propaganda and a sea of nonconsensual deepfaked porn.supported - garymarcus.substack.com
- Chatbots have been implicated in a number of suicides, and even mass murder.supported - garymarcus.substack.com
- AI developers should be obligated to demonstrate that the benefits for their new products outweigh the risks — and not have market access otherwise.supported - garymarcus.substack.com
- New York City is in a position to demand independent regulatory review for AI used in its borders.supported - garymarcus.substack.com
Caveats
- This is a claim by the speaker and not independently verified by the source.
- This is a claim by the speaker and not independently verified by the source. The source does not provide specific details or evidence for these implications.
- Single-source caution: verify critical details at the linked source.
Radar score 76/100 - how it was calculated
- Reliability 77: Primary on-record statement
- Freshness 100: Fresh source date
- Novelty 64: New statement by a tracked AI voice
- Technical 48: Structured technical source signals
- Developer 48: Builder relevance source signals
- Ecosystem 70: Tracked AI voice
- Confidence 98: Claims have reliable evidence
Discussion
Loading comments...