September 14, 2026
Jensen Huang told President Trump directly that Nvidia is "not going to let" an AI slowdown happen, according to TechCrunch. That's not a hedge or a diplomatic non-answer -- it's a flat rejection of the caution that Dario Amodei has been urging, and that even Elon Musk and Sam Altman have recently voiced some sympathy for. Huang's position isn't surprising once you follow the money: Nvidia doesn't sell caution, it sells GPUs, and every conversation about pacing AI development is, for Nvidia, a conversation about demand.
What makes this worth a business reader's attention isn't the disagreement itself -- disagreement among AI executives is now the norm -- it's who is doing the disagreeing. Amodei runs a frontier lab and has skin in the safety conversation. Huang runs the company that profits most directly from the industry never pausing. When the person supplying the picks and shovels says the gold rush isn't slowing down, that's a data point about where capital and compute are actually headed, regardless of what the model builders say in interviews. We've written before about the gap between what CEOs say publicly and what their companies do -- see our take on whether the 'slow down' crowd really means it -- and Huang's comment is the clearest evidence yet that the hardware layer of this industry has no interest in a pause, rhetorical or real.
For businesses evaluating AI tools right now, the practical takeaway is simple: don't wait for a slowdown that isn't coming. Model releases, price drops, and new capabilities will keep arriving on a compressed timeline, and the competitive gap between companies that adopt early and those that wait for things to "settle down" is likely to widen, not narrow. That said, Huang's confidence is also a reason for healthy skepticism -- a chip company's CEO has an obvious motive to talk up perpetual growth, and it would be naive to treat his prediction as neutral analysis rather than sales pitch dressed as forecast.
TechCrunch's own coverage today doubles as a reminder of where the industry's attention is going next: a session on using AI to help engineer species de-extinction, and a closing five-day window to grab an exhibit table at Disrupt 2026. Neither is breaking news in the way Huang's comment is, but together they show an industry still comfortable pouring resources into ambitious, headline-grabbing bets -- from bringing back extinct species to courting 10,000-plus founders and investors on an expo floor -- at the exact moment some of its most prominent voices are calling for restraint. We flagged this same tension in our earlier piece on why Disrupt matters even though it isn't AI news in the traditional sense: the events and the announcements around them tell you where money and attention are actually flowing, which is often a better signal than the press releases themselves.
None of this changes the calculus for a small or mid-size business trying to decide what to build or buy right now. If anything, it reinforces it. The infrastructure providers, the events industry, and the moonshot startups are all still moving at full speed. Waiting for consensus on AI safety before you adopt a tool is like waiting for consensus on the internet in 1998 -- it's a reasonable worry, but it's not a business strategy. Teams that want a lower-risk way to move fast can start with a workflow automation approach that keeps humans in the loop rather than betting everything on the newest frontier model.
Do you think Jensen Huang's confidence reflects where AI is actually headed, or is it just what you'd expect the industry's biggest hardware supplier to say?
Sources