Apple cuts daily AI spending limits for staff to curb rising costs

Tightening the AI purse strings

Apple is reportedly implementing stricter spending limits on the artificial intelligence tools its employees use. As the company pushes for wider adoption of AI across its workforce, it is simultaneously trying to rein in the high costs associated with these technologies. According to a report from Digit, the company recently reduced the daily spending limit for some engineers from $300 to $200.

The company is considering even deeper cuts to these allowances. For some engineering teams, the daily limit could be lowered to as little as $100. These restrictions are part of a broader effort to manage the financial impact of integrating AI into daily workflows while maintaining profitability.

The move has sparked a variety of reactions among the workforce. While some employees felt the original $300 limit was unnecessarily high since many did not use the full amount, others view these cuts as a sign of broader cost-cutting measures. Some staff members have expressed concern that these changes might precede wider organizational shifts or job cuts.

Understanding AI tokens and costs

To understand why these limits exist, it is important to understand how AI companies charge for their services. Most providers charge based on tokens, which are the basic units of text and data that an AI model processes. Every time an employee asks a question or gives a command, the system consumes a specific number of tokens to generate a response.

The complexity of a task directly correlates to the number of tokens used. For example, simple queries use fewer tokens, while complex tasks like writing or fixing software code require significantly more. Because more advanced models are more computationally intensive, they often cost more per token, making them more expensive for a company to provide to its staff on a large scale.

By capping the daily spending, Apple can better predict and control its operational expenses. The company is also encouraging staff to be more mindful of their choices. Employees are being asked to prioritize cheaper AI models or those that run directly on their devices, which typically cost less than cloud-based alternatives.

Impact on different departments

The spending limits do not affect every department in the same way. Currently, teams outside of the engineering department have a daily spending limit of around $100. However, Apple is considering reducing this amount to as little as $25 for some of those non-engineering teams.

This discrepancy suggests that the company views engineering tasks as more token-intensive. Because software development often requires the AI to process and generate large blocks of code, those teams require a higher baseline of spending. Non-engineering teams may be able to achieve their goals using simpler prompts that consume fewer resources.

Despite these restrictions, there is a mechanism for high-demand projects. Employees who can demonstrate that their specific work requires higher spending limits are still able to request an exemption. These requests must be reviewed and approved by senior leadership to ensure the extra spending is justified by the project's needs.

The broader tech landscape

Apple is not alone in trying to balance AI utility with its high price tag. Other major technology companies are implementing similar strategies to manage their internal costs. For instance, Microsoft has also tightened the internal spending limits for its own employees to keep costs manageable.

Other companies are using different methods to achieve similar goals. Tesla has introduced a $200 weekly limit for employee AI usage, while Uber has capped spending on individual AI coding tools at $1,500 per month. Meta has also introduced token budgets and is actively encouraging its staff to use its own internal AI tools instead of third-party ones.

These moves highlight a growing trend in the industry. As AI becomes a standard part of the corporate toolkit, companies are finding that they must be disciplined about how they deploy it. The goal is to find a sustainable balance between maximizing employee productivity and controlling the massive infrastructure costs required to power these models.

Rising infrastructure expenses

The need for these spending limits is underscored by Apple's own financial reports. The company reported a 32% rise in research and development expenses during fiscal year 2026. This significant jump was driven primarily by the higher costs associated with infrastructure and AI-related development.

As companies build out the hardware and software necessary to run large-scale AI, the overhead becomes a major line item. Even as Apple seeks to optimize its internal spending, the overall cost of the technology continues to climb. This creates a challenging environment where the company must innovate while simultaneously finding ways to save money.

Looking ahead, the company will likely continue to refine its AI strategy. This may involve a greater focus on optimizing existing models or developing more efficient on-device solutions. By moving tasks away from expensive cloud-based tokens and onto local hardware, Apple can potentially reduce its reliance on high-cost external providers.