Tesla Caps Engineer AI Budgets to $200/Week Over Soaring Token Bills

Internal memos sent to Tesla employees revealed a hard cap on AI tool expenses starting July 6. From now on, engineers get a weekly allowance of exactly two hundred dollars for querying external models. Any spend beyond this micro-budget requires direct managerial sign-off, ending the era of free-flowing corporate API keys.

The panic started when individual developers began racking up thousands of dollars a week in token costs. This financial cold shower is especially hilarious considering that Tesla spent the last six months aggressively pushing employees to use neural networks, even ranking them on public internal dashboards by how many tokens they burned through.

To make the medicine go down, Elon Musk left a sneaky backdoor open. The budget cap does not apply to beta versions of products from xAI, his own AI venture that builds Grok. It is a masterclass in corporate synergy: limit outside spending, and gently nudge your captive audience of engineers into your own ecosystem, even though reports show most Tesla coders actually prefer Claude by Anthropic.

Other giants have stumbled into this exact same token trap. Uber recently had to slam the brakes on employee AI spending after blowing through their entire annual budget in just four months, while Meta and Walmart quietly introduced similar guardrails. Meanwhile, Tesla is still planning to spend over twenty-five billion dollars on massive data centers, robotics, and autonomous driving.

Cutting a coder's weekly sandwich-money-sized AI budget while simultaneously dumping tens of billions into massive supercomputers is peak corporate logic. It turns out that building the sci-fi future is incredibly cheap, right up until the moment employees actually start using it.

Source: The Information