President Trump just announced an AI Action Plan for the United States, outlining how the administration will keep US companies at the cutting edge of artificial intelligence development. A key policy in that plan: building increasingly larger, more powerful datacentres. And who do you call for AI data centers? Nvidia, of course. Though this was news to President Trump at one point, who admitted during his speech at the AI summit in Washington D.C. that he'd never even heard of 'em: "What the hell is Nvidia?"
Speaking on a conversation he claims to have had in the past, Trump says, in a winding back-and-forth with an unknown other:
"I said, 'we'll break them up', that's going to be very hard, I said, 'why'. I said, 'what percentage of the market does he have?' 'Sir, he has 100%'. I said, 'who the hell is he?' 'What's his name?' 'His name is Jensen Huang, Nvidia.' I said, 'what the hell is Nvidia, I've never heard of it before.' He said, 'you don't want to know about it, sir.'"
Trump then reiterates that his initial thought was to break Nvidia up, to promote further competition.
"I figured we could go in, break 'em up a little bit, get 'em a little competition. I found out it's not easy in that business. I said, 'supposing we put the greatest minds together, they work end on end a couple years?' He said, 'No, it would take at least ten years to catch 'em, if he ran Nvidia totally incompetently from now on.' I said, 'alright, so let's go onto the next one.'"
Trump asks Nvidia boss Huang to stand up for a round of applause, calling him a "great guy" and stating "what a job you've done."
Throughout this, AMD's CEO Dr. Lisa Su is also in attendance. That might've been a tough watch for the red team's chief—not because Su and Huang are such bitter rivals they can't be in the room together—they seem to get along well enough—but more so Trump just proclaimed Nvidia to have total market dominance in AI chips and that no one could ever catch them, even if they had their worst decade ever.
But not to worry, Dr. Lisa Su was next on the list of praisees, getting a swift compliment from the US president.
"Lisa Su, from AMD, Lisa… Thank you, congratulations, great job."
Both CEOs would later go on to speak on the All-in Podcast, during which Jensen would praise Trump as "America's unique advantage".
The AI race has certainly propelled both companies' CEOs to a level of importance we couldn't quite have expected over the past decade. It wasn't that long ago that you could reasonably expect AMD's events to be largely made up of gaming announcements, rather than AI that dominates the company's every action today. Nvidia, at least, has been banging the enterprise drum for quite some time.
There's no doubt that Nvidia's CEO Huang has been walking a tightrope between Washington D.C. and Beijing in recent months, as the US had taken a stance of cutting off China's supply to powerful AI chips, which saw Nvidia lose out on a key market opportunity, and has since led to some awkwardness surrounding alleged chip smuggling, which Nvidia denies any involvement or knowledge of.
Today, Nvidia and AMD are both in better positions—the US is ramping up AI development with large-scale plans for AI, and both companies have been given the go-ahead to sell AI chips to China again. That's all well and good, but for how long? Is Huang walking away from this summit thinking 'I'm a great guy' or 'Trump nearly decided to break up my hard-earned empire on a whim'?
Perhaps Huang is thinking only 'how do I make enough GPUs for all this demand?'
OpenAI, xAI, Meta, and many more compute providers are fighting tooth and nail to build out AI capacity today. xAI and OpenAI have announced plans just this week for billions of dollars of investment in computing power, largely provided by Nvidia's chips. OpenAI with an expansion at its Texas 'Stargate' facility, planned to run with over 2 million chips, and xAI with another $12 billion it'd like to raise to buy/lease more GPUs.
Let's just hope all this demand for AI accelerators doesn't leave gamers with less and less supply in coming years. TSMC, the global manufacturing hub for most high-performance silicon today, only has so much room to spare. The Taiwanese company plans to expand at its central location in Taiwan and its more recent development in the United States.
1. Best overall: AMD Radeon RX 9070
2. Best value: AMD Radeon RX 9060 XT 16 GB
3. Best budget: Intel Arc B570
4. Best mid-range: Nvidia GeForce RTX 5070 Ti
5. Best high-end: Nvidia GeForce RTX 5090
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
