Abstract:
Apple has recently begun to exercise stricter control over the cost of using internal artificial intelligence tools. According to multiple media reports citing people familiar with the matter, Apple has reduced the AI usage budget of some employees in the hope of controlling growing operating expenses while continuing to advance its artificial intelligence strategy.
It is reported that Apple has reduced the AI usage quota of some engineering teams from US$300 to US$200 per day in October this year. At the same time, the company is also studying options to further reduce related expenses, and the daily limit of some engineering teams may be reduced to $100 in the future.
In addition to the engineering department, many non-engineering teams at Apple have previously implemented a daily usage cap of $100. According to reports, the company has discussed further reducing the quota for some teams to US$25 per day. For employees who really need higher quotas of AI resources to complete their work, Apple will allow them to apply for additional quotas from senior managers.
These budgets are mainly used to pay for the Token costs consumed by the generative artificial intelligence model. Token is the calculation unit used by AI models when processing text, code and other data. It is also an important billing basis for current mainstream AI services. Due to the extensive use of AI-assisted programming tools in the software development process, engineers usually have higher usage quotas.
Currently, Apple’s internal employees can use a variety of mainstream artificial intelligence services, including Anthropic’s Claude, OpenAI’s ChatGPT, and Google’s Gemini and other platforms. To further reduce costs, some teams have been encouraged to use cheaper models or 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 have continued to improve, the so-called "Token Maxxing" phenomenon has emerged in the Silicon Valley technology industry, in which employees use AI resources extensively to improve development efficiency. Some companies have even established internal rankings to compare employees' AI usage. However, as AI-related spending rises rapidly, more and more companies are beginning to re-evaluate whether these investments can generate sufficient returns in productivity.
Internal employees have had mixed reactions to Apple’s adjustments. Some employees believe that the previous daily limit of $300 was already quite loose, and many people never actually came close to using the upper limit, so the new regulations have limited impact on daily work. But others see this as part of a broader cost-control plan and are concerned about whether it will be accompanied by organizational changes or other cost-saving measures in the future.
Reports indicate that some of Apple’s business units have recently introduced external consultants to evaluate operational efficiency. Coupled with the tightening of AI budgets, some employees have speculated 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 R&D expenditures continue to grow. The company's latest financial data shows that R&D expenditures in fiscal year 2026 have increased significantly year-on-year, with artificial intelligence and infrastructure construction being one of the important factors driving cost growth. This means that Apple has not slowed down its AI strategic layout, but is seeking a more refined cost management method.
Analysts believe that Apple’s move reflects that the technology industry is entering a more rational stage of AI investment. After experiencing large-scale investments and rapid expansion in the early stages, companies began to pay more attention to actual rates of return and 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.
Comments