A hilariously ironic AI-generated video is making the rounds online, showing OpenAI CEO Sam Altman getting busted for stealing a GPU from a supermarket. In the clip shared on X, we see an Altman taking a GPU from a shelf at a store that looks like Target. He’s then accosted by a security guy, before protesting “Please, I really need this for Sora inference. This video is too good.” Funnier still is that the video appears to have been made by an OpenAI employee.
i have the most liked video on sora 2 right now, i will be enjoying this short moment while it lastscctv footage of sam stealing gpus at target for sora inference pic.twitter.com/B86qzUGlMqSeptember 30, 2025
The shared clip was made on Sora 2 and is currently one of the most popular videos on the combined AI generator and video-sharing app. Sora 2 is OpenAI’s new model, which launched this week, over a year after the company debuted its groundbreaking video generation model. It also turned its mobile app into a social media platform similar to TikTok, allowing creators to share their AI-generated videos.
The company claims that this is its most accurate and realistic model yet, with more available controls. You can see this in the video, with the only tell-tale sign that it’s an AI video being the one box of a GPU moving by itself after the digital Altman took the white box off the shelf. That and the awkward dialogue, of course.
The digital Altman stealing a GPU for OpenAI is ironic on a number of levels, especially as it brings memories of the time when OpenAI ran out of GPUs, delaying the release of GPT-4.5. The company also plans to get over a million GPUs by the end of 2025, with an end goal of 100 million. Clearly, the irony that OpenAI simply can't get its hands on enough graphics cards isn't lost, even on OpenAI's employees.
OpenAI is also moving from a non-profit into a for-profit model, with the company teaming up with several big tech companies (like Nvidia’s $100 billion investment) to help it hit a high-value IPO. However, all these investments are raising questions about OpenAI’s long-term viability, with the real Sam Altman not saying anything about it yet.
The video is also a reminder that the future of AI and generated videos could make deciphering fact and fiction more difficult than ever in the years to come. Although some might find this AI video funny, it also raises concerns about the authenticity of videos. After all, this AI sample, which looks as if it were recorded from a security camera, is quite realistic at first glance, and we can only expect AI models to get better moving forward. It highlights the need for stricter controls on aspects like the addition of metadata to all video content, making it easier for people to determine whether what they’re watching is real or AI-generated.
Follow Tom's Hardware on Google News, or add us as a preferred source, to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button!
