Rising AI Costs Push Uber, Microsoft and Apple to Tighten Employee Spending Limits!

Companies are reviewing their artificial intelligence (AI) spending as the cost of using AI tools increases, as per Business Standard news report.  

Higher employee usage has pushed some businesses to introduce spending limits, particularly for services that charge according to token consumption. 

Tech Companies Introduce Limits on AI Usage 

Uber exhausted its entire 2026 AI budget in April after encouraging engineers to use AI tools more frequently. Adoption of Anthropic’s Claude Code, introduced in December, rose from 32% in February to 84% in March. The company had to slow usage after spending exceeded its budget. 

Microsoft also reduced employee token spending in August and began monitoring consumption. Bloomberg reported that Apple was lowering its daily employee spending limit from $300 to $200, with a further reduction to $100 planned. 

Companies Reconsider Model Selection 

Businesses are increasingly using model orchestration to manage AI expenses. The system directs tasks to different models depending on their complexity, instead of relying on one large model for every requirement. 

Goldman Sachs, in its Connecting Dots report, said enterprises across technology, retail, telecommunications and financial services were placing limits on AI expenditure because of high costs. 

Smaller language models and open-source alternatives are being used for routine tasks, while expensive frontier models are reserved for work requiring greater reasoning and accuracy. Experts say smaller models trained on company data can reduce costs by up to tenfold while improving accuracy. 

Company Data Affects Token Usage 

Fractal Analytics co-founder and CEO Srikanth Velamakanni said AI models often lack information about a company's internal processes, terminology and data. 

Google Cloud's Andi Gutmans said Gemini Flash could handle many customer requirements without using the more expensive Gemini Pro. Google also uses a knowledge catalogue to give AI agents only the information required for a particular task. 

Infosys follows a similar approach. Its system first uses locally deployable open-source models and moves requests to more advanced models when the initial system cannot complete the task. 

For enterprises, controlling AI expenditure increasingly involves monitoring token usage and selecting models according to the work involved. 

Read More: Airlines Expand Premium Offerings as Demand Grows Beyond Economy Travel! 

Conclusion 

As AI adoption grows, companies are paying closer attention to the cost of running these systems. Spending limits and the use of smaller models are among the measures being introduced. 

Disclaimer: This blog has been written exclusively for educational purposes. The securities mentioned are only examples and not recommendations. This does not constitute a personal recommendation/investment advice. It does not aim to influence any individual or entity to make investment decisions. Recipients should conduct their own research and assessments to form an independent opinion about investment decisions. 

Investments in the securities market are subject to market risks, read all the related documents carefully before investing.