The Information reports that OpenAI bought tens of thousands of Mac mini and Mac Studio units. According to the account, the purchases were driven less by interest in Apple’s chips and more by the need for physical desktop machines running a real operating system: developers training computer-use agents require macOS environments at scale, and macOS is the only widely distributed desktop OS that ships preinstalled on consumer hardware.
Why physical Macs are being bought
AI research teams are increasingly building agents that interact with desktops — opening windows, clicking controls, moving files and automating applications. For many training workflows, developers prefer or require actual operating systems and hardware rather than pure emulation, so labs have been acquiring physical machines to give agents realistic environments.
Market context and numbers
The reports note this demand coincides with a 29% year-over-year increase in Apple’s Mac revenue, which reached $10.4 billion in the most recent quarter. The article also cites the price for the M6 Mac mini in China at 6,999 yuan, an increase of about 55% compared with the M4 launch price.
Effects on developers and the AI industry
Previously inexpensive machines — described in the report as $599 student-oriented models — have become the lowest-cost option for agent training rigs, and so have been added to procurement lists for AI labs. Some labs have raised billions to develop custom silicon and reduce reliance on GPUs, but the GPU remains on purchase orders; the Mac mini has simply been added alongside those buys. The report suggests that the increased mac demand can lead to higher retail prices for developers and users because research organizations are acquiring stock at scale.
Takeaway
In short, the need to teach agents to operate real desktop environments has created a sudden, large demand for Mac desktop hardware. That demand aligns with higher Mac revenue for Apple and notable price increases for the Mac mini, while AI labs continue to buy GPUs even as they invest in custom chips.
