“We expect that developing leading AI infrastructure will be a core advantage in developing the best AI models and product experiences,” said CFO Susan Li during an earnings call.
This commitment is now taking shape. On Monday, CEO Mark Zuckerberg unveiled Meta Compute, a new initiative aimed at strengthening the company’s AI infrastructure, alongside plans to substantially expand its energy footprint in the coming years.
Meta plans to expand its AI infrastructure significantly, targeting tens of gigawatts this decade and potentially hundreds over time. CEO Mark Zuckerberg emphasized that how the company engineers, invests in, and partners to build this capacity will become a key strategic advantage.
To lead the initiative, Meta appointed three executives: Santosh Janardhan, head of global infrastructure, will oversee technical architecture, software, silicon programs, developer productivity, and the company’s global datacenter network; Daniel Gross, co-founder of Safe Superintelligence, will manage long-term capacity strategy, supplier partnerships, and business planning; and Dina Powell McCormick, Meta’s president and vice chairman, will coordinate with governments on infrastructure deployment, investment, and financing.
The expansion reflects Meta’s commitment to supporting the energy-intensive demands of AI, with projections suggesting that U.S. electricity consumption could increase substantially in the coming decade.
There is a clear race among tech giants to develop cloud environments optimized for generative AI. Last year’s capital expenditure projections indicated that many of Meta’s peers are pursuing similar goals. Microsoft has actively partnered with AI infrastructure providers, while Alphabet, Google’s parent company, acquired data center firm Intersect in December. TechCrunch reached out to Meta for further details about its new initiative.
