Washington's AI playbook: fewer handcuffs, more chips
The Trump administration is heading into the G20 with a very clear message: don’t slap on a bunch of new AI regulators, and while you’re at it, please buy American tech. It’s basically “move fast and don’t invent another bureaucracy” — with Nvidia and the broader U.S. AI stack riding shotgun.
Nvidia isn’t just a chip company anymore
Jensen Huang showing up alongside OpenAI’s Sam Altman is a pretty loud signal that Nvidia wants to be seen as more than a fancy GPU factory. Washington is pitching chips, cloud infrastructure, models, and governance as a packaged export, which is a neat little way of saying: if the world builds AI, maybe it should build it on U.S. hardware.
That matters for investors because the company already said data center revenue hit $89 billion last quarter, up 117% year over year, and management pointed to sovereign customers as part of the boost. In other words, government-backed AI projects aren’t just policy theater — they can translate into more racks, more training runs, and more orders for Nvidia.
The global AI tug-of-war
Of course, the U.S. isn’t the only one trying to write the rules. China-backed efforts are pushing a different governance model, and Europe is tightening its own AI transparency requirements. So the fight isn’t just about who gets to regulate AI — it’s about whose chips, cloud, and software become the default operating system for the next tech wave.
Big picture: if the U.S. can keep AI regulation light abroad while exporting its full stack, Nvidia could keep turning geopolitics into revenue. Not a bad business model when the whole world wants the picks and shovels for the AI gold rush.