Announcing The Stargate Project
OpenAI
OpenAI just announced Stargate, a $500 billion plan to build massive AI data centers across the US. That's more money than most countries' GDP, all bet on needing way more compute.
Based on reporting by OpenAI — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
OpenAI wants to build the infrastructure backbone for whatever comes after GPT-4, and it's not being subtle about the price tag. The Stargate Project, unveiled alongside partners SoftBank, Oracle, and MGX, calls for $500 billion in AI infrastructure spending over the next four years, with $100 billion earmarked to start moving immediately.
The first physical site is already under construction in Abilene, Texas. Oracle handles the technical build-out, SoftBank brings financial muscle and oversees the business side, and MGX — the Abu Dhabi-backed investment vehicle — rounds out the money. Sam Altman framed this as the kind of compute buildout needed if you actually believe AI models keep getting more capable and more useful, rather than plateauing.
What's striking is the scale relative to anything OpenAI has done on its own. Half a trillion dollars for data centers puts this in the same conversation as national infrastructure projects, not typical corporate capex. It's a bet that the constraint on AI progress over the next few years isn't algorithms or talent — it's raw compute, chips, power, and the physical real estate to house all of it.
There's also a geopolitical layer here that's hard to ignore. MGX's involvement ties Gulf sovereign wealth directly into the buildout of American AI infrastructure, at a moment when governments everywhere are trying to figure out where they stand in the AI race. Stargate is as much a statement about who controls frontier compute as it is a construction project.
My take — AI-written commentary, not fact-checked reporting
Half a trillion dollars is the kind of number that should make everyone pause rather than nod along. This isn't proof AI is about to change everything — it's proof that a handful of companies and sovereign funds have decided to make that bet regardless, and the rest of us get to live with the power grids, land use, and market concentration that follow. I'd rather see that money split across a dozen smaller, more accountable efforts than funneled into one mega-project that only a few players can even participate in.
Read more about this at: OpenAI