Sam Altman tells Congress that AI regulation is “essential” but must let companies be flexible and adapt to new tech, recommends safety measures, and more
OpenAI's chief executive Sam Altman will tell US lawmakers on Tuesday that regulation of artificial intelligence must allow companies …
Financial TimesCristina Criddle
Context & Ripple Effects
Altman’s planned first congressional testimony on AI oversight had already put OpenAI at the center of the policy debate; the subsequent coverage records his case for flexible rules, safety measures, and a licensing regime in his congressional AI-regulation proposal.
The appearance was part of a broader Washington engagement that later included meetings with more than 100 members of Congress. It matters because a leading model developer was helping define the terms on which lawmakers might supervise the technology.
First-order effects
Congress gains a high-profile industry position: AI oversight should be real, but rules must accommodate rapid technical change and safety practices.
OpenAI is immediately positioned as a participant in shaping prospective AI governance, rather than solely a company subject to rules written without its input.
Second-order effects
Other AI developers and industry groups face pressure to articulate their own safety and compliance models as Congress weighs whether regulation should include a licensing approach.
A flexible, developer-informed framework could make engagement with federal policymakers a competitive necessity for companies building frontier AI systems.
Third-order effects
If this pattern persists, AI governance may develop through sustained negotiation between lawmakers and major model providers, making policy access and demonstrable safety processes part of industry structure.
The trend: This is an early data point in the rise of state-mediated AI governance, in which frontier labs seek to shape safety rules while preserving room to deploy new capabilities.
Boy oh boy, looking forward to Congresspeople trying to get their heads around AI https://www.ft.com/...
@pt
Parker on x
I assume this discussion will focus on big companies and rules that make sense for them, but I'm much more curious to hear how people are thinking about open source and rogue AI operators, who are likely to be less scrupulous and less controllable with regulation. https://twitter…
@garymarcus
Gary Marcus on x
The weight of it, the history of it, just hit me, hard. I will do my best to take my small part, tomorrow. [image]
@drkatedevlin
Kate Devlin on x
Gosh, it's almost like all the stuff loads of us have been saying for ages might get to happen because a tech bro has now said it. https://twitter.com/...
@alondra
Alondra Nelson on x
“Altman, whose company created AI chatbot ChatGPT, will say that ‘the regulation of AI is essential’” and “will recommend that AI companies adhere to an ‘appropriate set of safety requirements, including internal and external testing prior to release’” https://www.ft.com/...
@notbrunoagain
@notbrunoagain on x
Altman is pivoting from regulatory avoidance to regulatory capture. AI boosters shouldn't get any say in how their work is regulated. Or in anything, really. https://twitter.com/...
@mtrc
Mike Cook on x
Now that regulation is inevitable, the best next PR move if you own a big tech company is to lean into it and attempt to guide the process as much as possible. https://twitter.com/...
@emilymbender
@emilymbender on x
Hey Senate Judiciary Committee (@SenBlumenthal @mazieforhawaii @CoryBooker et al) are you actually interesting in learning about the risks of this tech? If so, you should be talking to @AJLUnited @timnitGebru @rajiinio @safiyanoble and def not Sam Altman https://www.judiciary.sen…
@ldrogen
Leigh Drogen on x
It was sadly inevitable that large firms would attempt to pull the ladder up on open source AI development Fuck these people https://twitter.com/...
@tcpeter
Tim Peter on x
Can you say, “pulling up the ladders”? https://twitter.com/...