Mark Zuckerberg says Meta is launching its own AI infrastructure initiative

In a significant move to enhance its artificial intelligence capabilities, Meta has officially launched a new initiative named Meta Compute. This announcement, made by CEO Mark Zuckerberg, underscores the company's commitment to expanding its AI infrastructure, a promise highlighted during last year's capital expenditure forecasts. Zuckerberg detailed the company's ambitious energy plans, stating that Meta aims to establish tens of gigawatts of energy capacity in this decade, potentially scaling up to hundreds of gigawatts over time. He emphasized that the strategic management of this infrastructure will be critical to gaining a competitive edge in developing top-tier AI models and user experiences. As a point of reference, one gigawatt equates to one billion watts, and the increasing demands of AI could see the United States' electrical consumption surge dramatically in the coming years. To spearhead the Meta Compute initiative, Zuckerberg has appointed three key executives. Santosh Janardhan, who has been with Meta since 2009 and serves as the head of global infrastructure, will oversee the technical architecture and the operation of the company’s extensive data center network. Joining him is Daniel Gross, a recent addition to Meta and co-founder of Safe Superintelligence, who will focus on long-term capacity strategies and industry analysis. Lastly, Dina Powell McCormick, a former government official, will collaborate with governmental bodies to facilitate the investment and development of Meta’s infrastructure. As the race intensifies to create AI-ready cloud environments, Meta's initiative reflects broader industry trends. Competitors like Microsoft and Google are also making significant investments in AI infrastructure, signaling a collective push towards advancing technological capabilities in the sector. TechCrunch has reached out to Meta for additional insights on this strategic initiative.