AI regulation – Jeff Bezos compared AI to a knife—an essential tool that can be misused, but one society shouldn’t ban or “outlaw” by shutting down data centers. Speaking on CNBC on Thursday, he argued regulation should focus on how AI tools are applied, not on removing the i
When Jeff Bezos talks about AI regulation, he starts with a kitchen tool—sharp, practical, and dangerous in the wrong hands.
“You don’t want to accidentally outlaw the knife because it can be used in a bad way. ” the Amazon cofounder said on CNBC on Thursday. “Knives are important tools and yes. every once in a while they get misused by someone. but you don’t say the solution to that isn’t to say. ‘OK. no more data centers. right?. No more knives. ’That’s not a smart approach to regulation.”.
Bezos said government regulation can serve “reasonable” purposes. and pointed to federal regulatory agencies like the FAA and the FDA as examples of how rules can protect public safety—whether that’s when people board planes or when they take prescription drugs. “There’s lots to be said for healthy government regulation to improve safety and products and so on,” he said. “And I don’t see why that won’t be applied at some point to the kinds of new tools that are being built by AI.”.
The line he drew is about balance. “You want to regulate the application level,” he added.
Those remarks land at a time when lawmakers and regulators are wrestling with how to govern a technology that has surged ahead faster than public trust and policy. Bezos did not dismiss the need for oversight. His argument was sharper: the answer shouldn’t be removing the systems that power AI.
The comparison to knives also came as Bezos formally stepped back into a CEO-style role at the same time he advises Amazon on AI. He is taking on his first CEO role since stepping down from his post as Amazon’s CEO in 2021. Bezos is serving as co-CEO of Prometheus. a physical AI startup he launched with Vikram “Vik” Bajaj. who helped create Google’s life sciences company. Verily.
Prometheus raised $12 billion in a Series B round. “That is a big chunk of the funding we’ve raised,” Bezos said. “And one of the reasons we’ve had to raise a significant amount of funding is because what we’re doing is very compute intensive.”
Bezos also pushed back on what the startup might become. “Addressing speculation about what Prometheus will do. Bezos said the startup is not building robots.” Instead. he said the company wants to build AI models that change the future of engineering and manufacturing. ultimately aiming for Bezos’ goal of creating an “artificial general engineer.”.
Bajaj, interviewed alongside Bezos, framed it in terms of engineering work getting faster and more capable. “It’s really a set of tools that will give those engineers the ability to turn their dreams into reality much, much more quickly than is possible,” he said.
The regulatory debate around AI isn’t sitting still. In the last year, backlash against the technology has sharpened the discussion, and AI companies have spent heavily on lobbying efforts as various states weigh potential regulations.
On the federal side. President Donald Trump signed an executive order allowing frontier AI model makers to voluntarily submit models for federal review up to 30 days before their public release. On Wednesday. Trump said he expects leading AI companies to agree to “giving back” to the public—comments that came after NOTUS reported that the White House is considering whether the US government should hold equity in AI firms.
Bezos did not directly address Trump’s order. Neither did Anthropic CEO Dario Amodei. who on Wednesday suggested that “models above a threshold of compute” should undergo mandatory third-party testing by the government or private organizations. Amodei also wrote in an essay that the government should have the power to block the release of AI models if such testing showed they would “present unacceptable risks.”.
Taken together, the conversations point to a central tension: how to control the most concerning uses of AI without turning the infrastructure that enables innovation into a collateral target. Bezos’s message was direct—don’t outlaw the tool simply because it can be misused.
Inside his own company’s push, the stakes are practical as well as political. Prometheus’ $12 billion Series B and its emphasis on being “very compute intensive” underscore why shutting down data centers would not be a neat fix. Bezos said the goal is different: regulate “the application level. ” while building the models that. in his view. will move engineering and manufacturing forward.
For now. the executive order. the lobbying push. and competing ideas about testing and possible release bans show how unsettled AI governance remains. And in Bezos’s framing. the question isn’t whether AI needs rules—it’s whether the rules will aim at misuse. or accidentally remove the knives themselves.
Jeff Bezos AI regulation data centers Prometheus Vikram Vik Bajaj Series B compute intensive engineering and manufacturing artificial general engineer Trump executive order FAA FDA third-party testing
So we’re just supposed to let them build infinite data centers? ok.
I don’t get the “don’t outlaw knives” thing. Like if it’s too dangerous, why not limit access? Data centers feel like the “problem” but maybe I’m missing it.
Bezos comparing AI to a knife is wild cause he’s basically saying don’t throttle the money machine. Also FDA/FAA?? those are totally different. People gonna get stuck with regulation on “application” and the rest is just loopholes.
“Regulate the application level” like what does that even mean? Like they gonna check every chatbot conversation like it’s a prescription? I thought the whole point was stopping the data centers from growing, not telling people to use it responsibly. But knowing regulators… they’ll do nothing until it’s too late anyway.