OpenAI’s CEO Sam Altman recently expressed a widely shared frustration in 2025: his company is currently “out of GPUs.” As someone who frequently searches for the best graphics card deals, I can relate. It’s indeed challenging to find these components these days.

Altman refers to high-end enterprise GPUs like the Nvidia DGX B200 and DGX H200 AI platforms—the latter of which OpenAI was the first to receive last year, personally delivered by Nvidia’s CEO Jen-Hsun Huang.

“GPT-4.5 is ready! Good news: it is the first model that feels like talking to a thoughtful person to me. I’ve had several moments where I’ve sat back in my chair and been astonished at getting actually good advice from an AI. Bad news: it is a giant, expensive model.” — February 27, 2025

As he elaborates, the company plans to incorporate “tens of thousands” of GPUs next week to introduce the new model, GPT-4.5, into its premium service. However, the ongoing GPU shortage has hindered the speed at which this can happen.

Altman furthers, “It is a giant, expensive model. We really wanted to launch it to Plus and Pro simultaneously, but we’ve been growing a lot and are out of GPUs. We will add tens of thousands of GPUs next week and roll it out to the Plus tier, but this isn’t how we want to operate; it’s hard to predict growth surges that lead to GPU shortages.”

Interestingly, it seems that OpenAI and Nvidia might face similar challenges in anticipating demand. The current environment has made acquiring GPUs quite difficult due to supply shortages and increased demand.

In an effort to secure its hardware future, OpenAI is reportedly considering developing its own AI chips to reduce dependence on Nvidia’s high-end silicon.

While Altman indicates that GPT-4.5 won’t necessarily set new performance records, he appears very proud of it:

“It is the first model that feels like talking to a thoughtful person to me. I’ve had several moments where I’ve sat back in my chair and been astonished at getting actually good advice from an AI.”

The tech community eagerly awaits the wider rollout, but perhaps if Nvidia miscalculates demand and sends an excess of GPUs, Altman might just find a few extra units heading his way.