Meta, historic agreement with Nvidia for millions of AI chips

The choice to strengthen the link with the Santa Clara supplier comes at a complex stage for Meta’s internal hardware strategy. Although the company has been working on proprietary solutions for running AI models for some time, technical difficulties and delays in the roadmap have made it necessary to rely on proven external partners. Nvidia, for its part, manages this leadership position by navigating between the fluctuations of the stock market and the pressure of increasingly fierce competition, represented by the recent agreements between AMD and companies of the caliber of OpenAI and Oracle.

The economic entity of the agreement remains confidential, but is part of a global investment framework that is redefining the parameters of the technology industry. The combined spending by Meta, Microsoft, Google and Amazon on AI infrastructure has now exceeded, in financial dimensions, the entire cost incurred by the United States for the moon missions of the 1960s. This production effort reflects the need to keep pace in a technological race where hardware optimization has become the determining variable for the sustainability of advanced digital services.