Tesla has put a weekly limit on how much money employees can spend on AI software, according to reports. The move to limit the amount of money employees can spend around AI software comes as companies around the world grapple with the huge costs of AI software and with the rise of generative machine learning.
According to a post on prediction market platform Kalshi on X, Tesla has decided to impose the spending limit on the company to control enterprise AI costs. While the company has not officially confirmed the policy, The Information says that the new policy will go into effect from July 6 and is based on an internal memo circulated among employees last month.
Some Tesla software engineers had been using AI services worth thousands of dollars in tokens per week, the report says, leading to tighter spending controls. Employees who need to exceed the $200 weekly limit need to get managerial approval before spending on AI-related expenses, the company says.
But that spending cap does not seem to apply to beta versions of AI products created by Elon Musk’s xAI, which points to Tesla’s continuing drive to promote the adoption of its own AI ecosystem.
As companies are getting AI incorporated into software development, research, customer support, content creation, and productivity workflows, subscription costs are becoming an increasing operational expense. OpenAI, Anthropic, Google, and xAI, for example, charge for premium AI models based on token usage or subscription rates, and hence enterprise-wide deployment can be far more expensive.
The report also says Tesla has been monitoring AI usage through internal dashboards ranking employees based on their use of AI token consumption. Tracking such users allows the company to identify heavy users and better manage software spending across teams.
The initiative is in line with CEO Elon Musk’s larger push to expand the use of xAI technology in Tesla. The company might need to work with internal AI models to help its employees reduce reliance on external AI providers and integrate with its own technology ecosystem, and it would be more efficient.
Earlier this year, Tesla introduced an internal AI platform called Bottle Rocket, which gives employees centralized access to all AI models from OpenAI, Anthropic, xAI, and Cursor. The platform also provides employees with unreleased AI models for testing and development, as well.
Several employees had been using personal AI accounts to do work related to Bottle Rocket prior to Bottle Rocket’s launch and had raised security, compliance, and cost-management issues before then. Tesla can track usage, enforce company policies, and reach enterprise agreements with AI providers via central AI access by centralizing access to AI services.
If true, Tesla’s spending cap also reflects an increasing trend in technology companies that aim to balance productivity gains of AI with financial discipline. As enterprise AI adoption accelerates, organizations are more and more focused on how to build AI capabilities and how they can use it at a lower cost and how to do so in a business manner.
Comments
Please to leave a comment on this article.