Google has imposed limits on Meta’s access to its Gemini AI models after Meta requested more computing capacity than Google could provide, according to a recent report citing sources familiar with the matter. The restrictions, reportedly communicated to Meta around March, have disrupted and delayed some of Meta’s internal artificial intelligence projects due to the shortfall in available capacity.

The demand for Google’s Gemini AI models has been particularly high from Meta, which has led to more significant impacts on the social media company compared to other Google clients, who have been affected to a lesser extent. In response to the capacity constraints, Meta has reportedly urged its employees to use AI tokens more efficiently—tokens being the units that measure AI usage.

Both Google, owned by Alphabet, and Meta have not publicly commented on the reports. Efforts to verify the information independently were not immediately successful.

The situation highlights broader challenges in the tech industry, as companies continue to invest heavily in chips and data centers to meet surging demand for AI services but still face limitations on computing power. Google Cloud’s recent financial disclosures illustrate this dynamic: despite revenue growth to $20 billion in the first quarter ending March, CEO Sundar Pichai noted that computing power constraints curtailed even greater expansion and contributed to the cloud unit’s backlog nearly doubling quarter over quarter.

These developments reflect the growing pressure on cloud providers and AI developers as they compete to scale infrastructure capable of supporting increasingly complex AI workloads. The restriction on Meta’s access to Gemini models underscores how supply limitations may affect innovation timelines and operational efficiency within major technology firms.