Tesla is putting a hard ceiling on how much its employees can spend on AI tools. Starting July 6, individual workers will be limited to $200 per week on third-party AI services, a move that comes after some software engineers reportedly ran up weekly token bills in the thousands of dollars.

The tab got too big

The new policy is a response to what happens when you tell engineers to adopt AI aggressively without putting guardrails on the budget. Some Tesla employees were burning through thousands of dollars a week in AI tool costs, mostly from third-party services like large language model APIs and coding assistants.

One notable carve-out: the spending cap will not apply to beta versions of xAI products. That’s Musk’s own AI company, the one behind the Grok chatbot. So Tesla is essentially telling employees they can use as much in-house AI as they want, but the third-party buffet now has a price limit.

Tesla isn’t alone in the AI austerity era

Uber provides maybe the most dramatic cautionary tale. The ride-hailing company reportedly burned through its entire annual AI budget in just four months, forcing it to institute a $1,500 monthly cap per tool per employee.

The spending paradox

At the same time Tesla is limiting individual employee AI spending, the company has dramatically increased its corporate-level AI investment. Tesla raised its 2026 capital expenditure guidance to over $25 billion, nearly tripling the previous year’s allocation.

That money is being directed toward computing infrastructure, robotics, and autonomous driving technology. A $200 weekly cap on employee tool spending is a rounding error compared to a $25 billion-plus capex budget.

The xAI exemption reinforces this reading. By letting employees use Musk’s own AI products without limits while capping competitors, Tesla is simultaneously controlling costs and funneling usage toward its own ecosystem.

What this means for investors

The pattern of companies capping AI spending while simultaneously increasing infrastructure investment tells a clear story about where the smart money is going. Tesla’s xAI exemption is the most explicit version of this trend.

The fact that multiple large companies, Tesla and Uber among them, are all hitting similar walls around AI costs suggests this isn’t a company-specific problem. Usage-based pricing made sense when adoption was limited. At enterprise scale, it becomes a budget black hole.