OpenAI's Sam Altman says people shouldn't run AI on their own computers
He argues running AI locally wastes limited computer parts — a take that has sparked anger online.

Altman made the remarks on the Cheeky Pint podcast 1. He argued that the push for "local inference" — running AI models directly on your own device rather than through a cloud service — doesn't make sense when computer chip supplies are tight .
The chip shortage and why he objects
The AI boom has made individual computer components, especially memory and storage, extremely expensive. US chipmaker Intel saw its shares plummet nearly 30% on a Friday, marking the worst stock collapse in 52 years, which analysts linked to a major slowdown in the PC market 1.
Altman blamed part of the problem on individuals wanting to run AI models locally instead of leaving the remaining chips for large data centers 1.
Many individuals are interested in running models locally, yet they mistakenly believe the issue is already resolved since open-source models will be available. They think we will be able to run them on personal computers and therefore won't require the infrastructure found in data centers.
He added: "But from a global allocation perspective, actually, it's quite terrible to have everybody have a bunch of the world's limited supply of dynamic RAM sitting in their own computer" 1.
Cloud vs. home computing
Instead of running AI at home, Altman argued that cloud services operated by massive companies like OpenAI are the better solution 1.
"You want maximum utilization in the cloud if you're trying to deal with this very constrained environment," he said. "And somehow everyone's, many people's intuition is the opposite of that" 1.
He also framed it as a shared responsibility: "Given that the world is going to be limited in its silicon capacity, we all have a shared interest" 1.
The trend toward local AI has grown thanks to open-weight models coming from China, which allow people to download and run AI software on personal machines 1.
Guess
How far ahead are PC makers fully booked for circuit boards?
PC makers are booked through the end of 2030 due to a massive circuit-board shortage.
Backlash from tech enthusiasts
The comments drew immediate fury, particularly from the PC gaming community 1.
One Reddit user called Altman a "self-centered POS" and criticized his "cult leader mentality" 1. Another wrote that Altman seems unable to understand why people want to own their own computers, calling it an "alien concept" to him . A third commenter argued that privately owned companies driven by profit should not be making monumental decisions affecting the vast majority of the population .
The backlash has also been fueled by Altman's previous remarks this month, where he argued that the "world should accept some bad things happening for the benefits of this technology" — comments that also drew heated responses 1.
The controversy comes at a tense time for the AI industry, which continues to struggle with commercializing its products while expenses balloon and investors grow nervous 1. The idea that people could run AI locally on their own machines undermines the business model of expensive cloud services offered by companies like OpenAI .
There is also a massive shortage of circuit boards, with PC makers already fully booked through the end of 2030 1.
Key facts
- Sam Altman opposed local AI inference on personal computers
- Intel shares dropped nearly 30%, worst collapse in 52 years
- Altman argued cloud services are more efficient use of limited chips
- PC gaming community responded with fury on Reddit
- Open-weight models from China have fueled the local AI trend
Sources
Keep reading
New modelsOpenAI launches Decisions API for fast yes no AI evaluationsHerald of AINew modelsOpenAI launches Decisions API for fast yes no AI evaluations
A new API turns complex text and image analysis into simple answers — and runs about ten times faster than the Responses API.





