Apple Cuts Internal AI Usage Budgets to Reduce Spending as Costs Rise

Apple is reducing the amount employees can spend on artificial intelligence tools as part of a broader effort to control operating expenses, even as the company encourages greater AI adoption across its workforce. According to a Bloomberg report published on November 4, 2026, Apple lowered the daily AI spending allowance for some engineers from $300 to $200 in October and is considering additional reductions that could bring the limit down to $100 for certain engineering teams.

The changes highlight a growing challenge for major technology companies: balancing the productivity benefits of generative AI against the cost of running increasingly sophisticated models. Apple employees use tools including OpenAI’s ChatGPT, Anthropic’s Claude and Google’s Gemini for coding and other work-related tasks. However, rising token consumption and infrastructure expenses are prompting companies to examine how much employees spend on AI services and whether cheaper alternatives can deliver similar results.

Apple Reduces Daily AI Spending Limits for Employees

Apple’s initial reduction affected some engineering teams, whose daily AI spending allowance was lowered from $300 to $200, according to Bloomberg.

The company is now considering further reductions for certain engineering groups, potentially bringing their daily allowances to $100.

Employees outside engineering already face a daily spending limit of around $100 in many groups. Apple is reportedly considering lowering the allowance for some of these employees to as little as $25.

The restrictions are not necessarily uniform across the company. Employees who require additional computing resources for their work may be able to request exemptions from senior executives.

The reported changes suggest Apple is moving towards a more controlled approach to internal AI usage rather than allowing employees to experiment with expensive models without close spending oversight.

Employee category	Earlier daily limit	Reported or proposed limit
Some engineering teams	$300	$200
Certain engineering teams	$300	Potentially $100
Many non-engineering groups	$100	Potentially $25

These figures represent reported spending allowances, not a confirmed company-wide reduction affecting every Apple employee.

Why Apple Is Cutting AI Budgets

The primary reason behind the restrictions is the cost of using generative AI models.

Most commercial AI services charge customers according to token consumption. Tokens are small units of text or other information that AI models process when responding to requests.

The more information a model processes, the higher the potential cost. Complex coding tasks can be particularly expensive because developers may repeatedly ask AI systems to generate code, inspect large repositories, identify errors and suggest corrections.

For example, an engineer working with an advanced AI coding assistant may generate thousands of lines of code, analyse multiple files and run repeated debugging sessions. These activities can consume substantial quantities of tokens.

More capable models can also carry higher usage costs than smaller alternatives.

Apple’s decision appears aimed at ensuring that employees use AI resources more efficiently while maintaining access for tasks that genuinely require expensive models.

Apple Employees Use ChatGPT, Claude and Gemini

Apple employees reportedly have access to multiple external AI services, including:

• <Entity value=”ChatGPT” category=”software” disambig=”OpenAI conversational AI assistant”/>: Used for writing, reasoning, research and coding assistance.
• <Entity value=”Claude” category=”software” disambig=”Anthropic AI assistant and coding models”/>: Used for software development and complex knowledge work.
• <Entity value=”Gemini” category=”software” disambig=”Google generative AI models and assistant”/>: Google’s AI platform for reasoning, research and productivity tasks.

Giving employees access to multiple models allows teams to select tools based on the complexity of their work.

However, using several external AI providers can make spending difficult to manage. Different models have different pricing structures, context limits and capabilities.

Apple is reportedly encouraging some teams to switch to less expensive models or use AI systems that run locally on company devices.

On-device AI can reduce dependence on external cloud services for certain workloads, although local models also require hardware resources and may not match the capabilities of the most advanced cloud-based systems.

Apple’s AI Spending Is Rising Despite Cost Controls

The reductions come at a time when Apple’s overall spending is increasing.

For fiscal year 2026, Apple reported a 32% increase in research and development expenses, reflecting higher investment in infrastructure and AI-related activities.

Companywide operating expenses also increased 22% to approximately $75.5 billion, representing around 16% of net sales.

These figures illustrate the financial pressure facing Apple as it invests in new technologies while attempting to maintain operating efficiency.

The company is working to expand AI capabilities across its products and internal operations, but it has taken a relatively measured approach compared with competitors investing heavily in developing and operating large foundation models.

Rather than building every AI capability independently, Apple has also explored partnerships with external AI providers.

The reported employee spending restrictions indicate that even companies with substantial financial resources are examining the economics of everyday AI usage.

Apple’s New CEO Pushes for a Leaner Organisation

The AI budget reductions also come during a broader organisational efficiency drive under CEO <Entity value=”John Ternus” category=”people” disambig=”Apple chief executive officer”/>.

Ternus took over as Apple’s chief executive in September 2026 after succeeding Tim Cook.

Bloomberg has reported that Ternus wants Apple to operate with fewer management layers and a more streamlined organisational structure.

Some Apple divisions have also reportedly hired external productivity consultants to examine operations and identify opportunities to improve efficiency.

These developments have raised questions among employees about possible restructuring and future workforce changes. However, the reported AI budget reductions do not establish that Apple has announced layoffs connected to the spending changes.

The company appears to be reviewing how resources are allocated across different teams while attempting to improve productivity.

Big Tech Companies Are Also Restricting AI Spending

Apple is not alone in reassessing the cost of employee AI usage.

Technology companies initially encouraged widespread experimentation with generative AI tools, hoping that employees would discover ways to improve productivity and accelerate software development.

However, unrestricted usage can create unpredictable bills, particularly when employees use advanced models for lengthy coding and reasoning tasks.

Other major companies have introduced their own controls.

Microsoft has tightened internal AI spending allowances, while Tesla has reportedly imposed weekly spending limits on employee AI usage. Uber has also restricted the use of individual AI coding tools through monthly budgets.

Meta has introduced token budgets and encouraged employees to use internally developed AI tools.

These decisions indicate a shift from encouraging employees to use as much AI as possible towards measuring whether the resulting productivity improvements justify the costs.

Can Lower AI Budgets Affect Employee Productivity?

Restricting AI spending could create trade-offs for engineering teams.

Advanced AI coding tools can help developers write software, investigate bugs, understand unfamiliar codebases and automate repetitive work. Limiting access to these tools could reduce productivity if employees regularly reach their spending caps.

However, not every task requires the most expensive AI model.

Smaller models can often handle routine coding assistance, summarisation and basic research at lower costs. Companies can therefore reduce spending by directing simple tasks towards cheaper systems and reserving advanced models for complex work.

Apple’s reported approach reflects this distinction.

The actual financial savings remain uncertain because some employees reportedly did not use their previous $300 daily allowances in full. Nevertheless, tighter limits could become more significant if AI usage continues to grow.

The Bigger Picture

Apple’s decision reflects a broader change in the economics of enterprise AI. Companies initially focused on expanding access to generative AI, but the rapid growth in token consumption and computing expenses is making cost management increasingly important.

The challenge is no longer simply giving employees access to powerful AI models. Businesses must determine which tasks benefit from expensive models, which can be handled by smaller systems and whether the productivity gains justify the additional expenditure.

For Apple, this is particularly important as it increases investment in AI capabilities while maintaining its focus on operating efficiency. The company’s approach could also influence how other large employers manage AI access across engineering, research and administrative teams.

Looking Ahead

Apple’s next challenge will be finding the right balance between AI spending controls and employee productivity. If tighter budgets push teams towards cheaper models without affecting output, the company could reduce unnecessary expenditure while retaining the benefits of AI-assisted work. However, overly restrictive limits could slow developers who depend on advanced models for complex projects.

More broadly, the technology industry is entering a phase in which AI adoption will increasingly be judged by measurable returns rather than usage levels alone. As companies gain more experience with generative AI, token budgets, model selection and on-device processing are likely to become important parts of enterprise technology spending strategies.