(Bloomberg) — Google () imposed limits on Meta Platforms Inc. ()’s use of its Gemini artificial intelligence models because it was unable to provide as much computing power as the social media company wanted, according to the Financial Times.
In the latest sign of AI infrastructure restrictions, the Alphabet Inc. search giant has applied restrictions to several customers, with Meta being particularly affected, the report said. FT this Sunday.
The move had ripple effects on Meta’s internal projects and led the company to direct its employees to make more efficient use of AI tokens, according to the newspaper, citing three unidentified people familiar with the matter.
InfoMoney Tool
Download now (and for free)!
Google and Meta declined to comment to the FT.
Meta had initially turned to Gemini, which had proven superior to its own open-source Llama models, to automate security processes such as removing harmful content and eliminating scams, the paper said. However, the company has been increasingly using its new Muse Spark model as it seeks to reduce reliance on external models.
The AI boom is testing the limits of the computing capacity and energy needed to power the profusion of data centers that are emerging. In early June, Google agreed to pay Elon Musk’s SpaceX $920 million a month for computing capacity as part of a $30 billion cloud services contract running until mid-2029 as it races to meet voracious demand for capacity.
Meta, which doesn’t sell cloud computing services, is elevating spending on AI as CEO Mark Zuckerberg’s top priority, with the technology now at the center of his corporate vision.
Earlier this year, the social media company informed employees of plans to cut 10% of its workforce, or 8,000 positions, in an effort to offset its heavy spending. It also reassigned 7,000 employees to new AI-related roles as part of a broader corporate restructuring.
© 2026 Bloomberg L.P.
Continues after advertising
