Apple reportedly cuts some engineers’ daily AI budgets from $300 to $200

Some Apple engineers now have $200 a day to spend on AI, down from $300 in October, according to Bloomberg reporting summarized by Investing.com on November 4. Further reductions are under consideration, alongside instructions for some teams to use cheaper models or AI that runs directly on devices. Employees needing more could seek executive approval.

The allowance pays for tokens

The daily budgets cover tokens: units of text and other data that AI models process when handling requests. Providers typically charge according to how many tokens a task consumes. The allowance therefore limits spending on that processing, rather than setting a fixed number of questions an employee can ask or coding tasks they can complete.

Model choice also affects the bill. More advanced models generally cost more, while complex jobs such as writing and debugging software can consume large quantities of tokens. Engineers have the largest allowances because coding can be token-intensive.

Apple employees have access to services including Anthropic’s Claude, OpenAI’s ChatGPT and Google’s Gemini.

Lower caps remain under consideration

• Some engineering teams could see their daily allowances fall to as little as $100.
• Many groups outside engineering already have a $100 daily cap. Apple has considered reducing the allowance to as little as $25 for some teams; that lower figure remains a possibility rather than an implemented limit.

A smaller ceiling is not measured savings

Some employees told Bloomberg that few workers were using their full $300 daily allowance, according to MacRumors’ summary. They viewed the old ceiling as unnecessarily generous. The amount Apple will save remains unclear.

The restrictions come against a larger increase in Apple’s costs. The company said research and development expenses rose 32% in fiscal 2026, driven by higher infrastructure and AI costs. That figure covers the research and development expense category, rather than isolating employees’ use of outside AI models.

Companywide operating expenses increased 22% to $75.5 billion, equal to 16% of net sales.