Tesla Pumps Brakes on In

Tesla Pumps Brakes on In-House AI Usage

Tesla has reportedly become the latest major company to restrict employee artificial intelligence usage.

The electric car company is set to impose a $200 weekly limit on staff’s AI spending beginning Monday (July 6), The Information reported, citing an internal memo.

The report calls the cap a sign that even companies focused on using AI to reshape their operations and products are becoming more cost conscious.

In the last few months, Tesla software engineers were often consuming thousands of dollars’ worth of tokens each week, the report added, citing two sources familiar with the usage. Now, those sources added, workers will need permission to go above the new limit.

PYMNTS has contacted Tesla for comment but has not yet gotten a reply.

As the report noted, what is happening at Tesla has happened at several other companies — including Meta, Uber and Walmart — who have gone from pushing workers to up their AI usage to placing curbs on their spending.

The financial friction these companies hit “traces back to a structural mismatch between how AI tools are priced and how enterprise finance teams are built,” PYMNTS wrote in May.

As covered here, annual licenses and seat-based pricing gave CFOs a stable cost structure they could forecast. Token-based consumption, where charges add up based on volume of text processed and generated, broke that model open.

“A surge in internal experimentation, a new product feature or a poorly optimized prompt can cause costs to spike in ways that are difficult to anticipate,” PYMNTS added.

The Information report says the speed of how this has played out at Tesla is notable, because the company has lagged behind other tech/corporate giants in formalizing workers’ AI usage.

The report also points to the stakes facing Tesla’s AI rollout, as it is playing out against the company’s wider vision of using AI in its products. Tesla leader Elon Musk has said the company’s future hinges on whether it can deploy AI in its robotics/robotaxi business amid several quarters of flagging revenues.

Writing about limits on AI usage in another May report, PYMNTS noted that AI access is moving from an “always-on utility” to a managed service shaped by pricing tiers, limits and usage windows. For users, that means adjusting behavior around those limits, not expecting continuous access.

“For providers, it turns infrastructure efficiency and cost control into competitive levers that directly shape adoption,” the report added.

“As demand continues to rise, the companies that can balance performance with more generous, predictable access will have an edge, while others risk losing users to friction that becomes increasingly hard to ignore.”

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.