Apple tightens internal AI spending, cutting some engineers' daily quota from $300 to $200

Reports say that in October of this year, Apple already lowered the AI usage quota for some engineering teams from $300 per day to $200. At the same time, the company is also studying plans to further compress related expenses, and the future daily limit for some engineering teams may drop to $100.

In addition to the engineering department, many non-engineering teams at Apple had previously already implemented a daily usage cap of $100. The report says the company has discussed further compressing the quota for some teams to $25 per day. For employees who genuinely need a higher AI resource quota to complete their work, Apple will allow them to apply to senior management for additional quota.

These budgets are mainly used to pay the token costs consumed by generative artificial intelligence models. Tokens are the computational units used by AI models when processing text, code, and other data, and are also an important billing basis for mainstream AI services today. Because AI-assisted programming tools are heavily used in the software development process, engineers usually have higher usage quotas.

At present, Apple's internal employees can use a variety of mainstream artificial intelligence services, including platforms such as Anthropic's Claude, OpenAI's ChatGPT, and Google's Gemini. To further reduce costs, some teams have been encouraged to use lower-priced models, or to prioritize AI solutions that can run directly on local devices.

It is understood that over the past year, as the capabilities of AI programming assistants continued to improve, a phenomenon known as "Token Maxxing" emerged in the Silicon Valley tech industry, in which employees make extensive use of AI resources to improve development efficiency, and some companies even set up internal leaderboards to compare employees' AI usage. However, as AI-related spending has risen rapidly, more and more companies have begun to reassess whether these investments can generate sufficient productivity returns.

Regarding Apple's adjustment this time, internal employee reactions have been mixed. Some employees believe that the previous quota of $300 per day was already quite generous, and many people had in fact never come close to the usage limit, so the new rules have limited impact on daily work. But others see this as part of a broader cost-control plan and worry whether it will be accompanied by organizational adjustments or other cost-saving measures in the future.

The report says that some of Apple's business departments have recently also brought in external consultants to evaluate operational efficiency. Combined with the tightening of AI budgets, this has sparked speculation among some employees about future restructuring plans. However, there is currently no evidence that Apple has formulated relevant layoff plans.

It is worth noting that Apple's overall research and development spending is still continuing to grow. The company's latest financial data show that R&D expenses in fiscal year 2026 increased significantly year over year, among which artificial intelligence and infrastructure construction are among the important factors driving cost growth. This means that Apple has not slowed down its AI strategic layout, but is seeking a more refined approach to cost management.

Analysts believe that Apple's move reflects that the tech industry is entering a stage of more rational AI investment. After early large-scale investment and rapid expansion, companies have begun to pay more attention to actual returns and to establish stricter management mechanisms for the use of internal AI resources. While continuing to promote the integration of artificial intelligence into products and business processes, how to control high operating costs is becoming a new issue faced by the entire industry.