According to The Information, two informed sources revealed that Meta had agreed to invest over $10 billion over the next six years in Google Cloud Services, focusing on AI infrastructure. This marked the first time that Meta was working with Google Clouds on a large scale to supplement its own data centre.
At the heart of the agreement was support for Meta ‘ s AI infrastructure. Meta is significantly expanding its Llama large-language model family and integrating AI technology into its full-line products (e.g. generation AI tools, advertising optimization, meta-cosm-related applications, etc.). The high performance computing (HPC) resources provided by cloud computing, in particular the GPU clusters, are essential for the training and deployment of large AI models.
Google is following up with the leading Amazon AWS and Microsoft Azure through a large cloud contract. Earlier this year, Google won the cloud business from OpenAI, which relies heavily on Azure. Data from Alphabet in July show that the Google Clouds sector (infrastructure and productivity software subscriptions) received $13.6 billion in the second quarter, with operating profits of $2.83 billion, and that 32 per cent of the increase in revenue exceeded 13.8 per cent for the company as a whole.
According to sources, Meta’s collaboration with Google is mainly centred on artificial intelligence infrastructure. Meta expects total expenditures to be as high as $114-11.8 billion in 2025, mostly for AI infrastructure and talent recruitment to expand its Llama model family and integrate AI into services. The choice of Google Clouds reflects Meta ‘ s pressing need for computing, especially in the longer-term self-built data centre cycle.
While Meta has long been competing with Google in the field of online advertising, Meta urgently needs access to all available cloud infrastructure. In addition to its own data centre, the company has committed to using Amazon and Microsoft cloud services. This collaboration with Google Clouds resulted in Meta becoming a co-client of the world ‘ s three main cloud service providers (AWS, Azure, Google Cloud) and demonstrated its diversified strategy in cloud computing.
The explosive growth of AI applications (generated AI, large-language models, referral systems, etc.) contributed to the need for cloud computing resources. The surge in demand for GPU, TPU and high bandwidth storage has made cloud suppliers the backbone of the AI competition. Meta chose Google Clouds to show a combination of considerations of the technical capabilities and price competitiveness of suppliers.