Apple Cuts Engineers’ Daily AI Budget From $300 to $200 as Token Bills Bite
Apple has cut the daily AI spending allowance for some engineers from $300 to $200, Bloomberg reported, with deeper cuts to $100 — and as low as $25 outside engineering — under consideration as token bills come due.
Apple is quietly rationing its own staff’s use of artificial intelligence. The company cut the daily AI spending allowance for some of its engineering teams from $300 to $200 in October, Bloomberg reported, in a move first summarised by TechCrunch, 4 November 2026. More cuts are reportedly being prepared: some engineering allowances could fall to $100 a day, and outside engineering the company is weighing a daily limit as low as $25.
Highlights
- Apple cut its engineers’ daily AI spending cap from $300 to $200 in October, Bloomberg reported via TechCrunch.
- Further cuts to $100 a day are reported to be in the works for some engineering teams.
- Outside engineering, a daily limit as low as $25 is reportedly being weighed.
- The allowances pay for AI tokens across tools such as ChatGPT, Claude and Gemini.
- TechCrunch said it has reached out to Apple for comment.
The budgets are not petty cash. They pay for the AI tokens engineers burn through when they use models such as ChatGPT, Claude and Gemini in their daily work, according to the Bloomberg reporting cited by TechCrunch, 4 November 2026. At heavy usage, those tokens add up quickly, which is why a daily cap is a direct lever on cost rather than a symbolic one.
From tokenmaxxing to token rationing
The cut reverses the mood of earlier this year, when, as TechCrunch, 4 November 2026 put it, a “tsunami of tokenmaxxing” took hold and companies pushed staff to use AI as much as possible in the name of vaguely defined efficiency gains. That frenzy was short-lived: once the token bills came due, rationing began to look like the more fiscally sensible approach. Concerns persist across the industry about the gap between what AI costs to build and run and the often unclear productivity gains it delivers.
The timing is striking because the wider debate about AI’s pace is intensifying at the same time. As our coverage of Anthropic’s warning that AI could soon speed up AI development beyond evaluation shows, labs are simultaneously worried about capability running ahead of oversight — and, as Apple’s cap shows, about usage running ahead of budgets.
Restraint inside, expansion outside
The internal restraint contrasts with how aggressively Apple is still pushing AI towards customers. The company is preparing to widen its consumer AI offer further, with Apple Intelligence heading to China with iOS 27.3 in December. In other words, AI for users remains a growth story worth funding, while AI for employees is now being metered like any other utility.
What to watch
Three questions follow. First, whether the deeper tiers — $100 for some engineering teams and $25 elsewhere — actually land, which would signal that October’s cut was the start of a programme rather than a one-off adjustment. Second, whether other large employers copy the model: per-employee token allowances are an easy policy to imitate once one major company normalises them. Third, whether the caps change how engineers work, pushing them towards cheaper models and more deliberate prompts. Apple has not publicly commented; TechCrunch, 4 November 2026 reported that it has reached out to the company for comment, and none was included in its report.
