Google reportedly limits Meta's access to Gemini AI models due to capacity constraints

According to a Financial Times report, Google has imposed restrictions on Meta's ability to utilize its Gemini artificial intelligence models. This decision, conveyed to Meta around March, was necessitated by Google's inability to provide the extensive computing capacity that the social media company had sought. The resulting shortfall in resources has reportedly led to significant disruptions and delays in various internal AI projects at Meta. While other clients of Google have also encountered similar limitations, Meta's exceptionally high demand for Google's AI models has made it uniquely impacted by these restrictions. This scenario highlights a broader industry-wide challenge: despite substantial investments in chips and data centers, technology companies are finding it increasingly difficult to secure enough computing power to keep pace with the escalating demand for AI services. Google Cloud, the company's cloud computing arm, recorded revenues of $20 billion in the first quarter ending in March. However, Google CEO Sundar Pichai noted that constraints in computing power hindered further growth and contributed to a near doubling of the cloud unit's backlog compared to the previous quarter. In light of these limitations, Meta has reportedly advised its staff to be more efficient in their consumption of AI tokens, which serve as a measure of AI usage. Neither Google nor Meta immediately responded to requests for comment outside of standard business hours. This development underscores the intense competition for advanced AI resources and the strategic importance of controlling access to both cutting-edge AI models and the underlying infrastructure required to operate them.