What We Learned About the Future of AI from Microsoft, Meta Earnings

Artificial intelligence was the focus on Wednesday when tech giants Microsoft (MSFT) and Meta (META) kicked off the first round of Big Tech earnings of 2025.

Analysts peppered executives with questions about their AI investments, how much AI is contributing to revenue, and their thoughts on Chinese start-up DeepSeek, which shook up Wall Street earlier this week with its super-efficient AI model.

AI Business Is Booming

Executives were eager to show investors that artificial intelligence is contributing to their bottom lines.

Meta CFO Susan Li said Meta AI reached 700 million monthly active users in the fourth quarter while the company’s AI-powered Advantage+ marketing service grew 70%. Meta’s quarterly revenue and earnings both surpassed analysts’ expectations.

Microsoft’s fiscal second-quarter cloud revenue fell slightly short of expectations, but CEO Satya Nadella noted the company’s AI revenue run rate hit $13 billion in the quarter, well above the $10 billion mark he forecast in October.

IBM (IBM) CEO Arvind Krishna said the company booked nearly $2 billion of AI business in the final quarter of 2024; that accounts for about 40% of the $5 billion booked since inception. 

Tech Titans Are Committed to Their AI Spending

Executives from Microsoft and Meta stood by their plans to spend tens of billions of dollars this year on AI infrastructure.

“We expect to continue investing against strong demand signals,” Microsoft CFO Amy Hood said of its capital expenditures plan on the company’s earnings call. “However, the growth rate will be lower than FY2025.” 

Microsoft forecast spending $80 billion on infrastructure in its 2025 fiscal year. So far, two quarters in, the company has spent $30.7 billion, a 56% increase from a year ago. 

Meta CEO Mark Zuckerberg also defended his company’s massive AI investments, which could total $65 billion this year. “I continue to think that investing very heavily in CapEx and infra is going to be a strategic advantage over time,” he told analysts on Wednesday evening. 

Big tech’s AI spending was already a source of anxiety on Wall Street, where for half a year questions have swirled about whether demand will match the scale of infrastructure investments. Those doubts were amplified this week when markets took notice of DeepSeek, whose open-source AI model reportedly performs as well as leading U.S. models and runs at a fraction of the cost. 

Heading into Wednesday’s earnings reports, analysts doubted that hyperscalers like Microsoft and Amazon (AMZN) would change their investment outlooks. DeepSeek’s efficiencies could fuel the development of AI applications and stoke demand further, giving cloud providers even more reason to continue aggressively adding computing capacity. 

Silicon Valley Impressed by DeepSeek, Not Scared

Markets may have been shaken by DeepSeek, but tech executives weren’t. 

“In some sense, what’s happening with AI is no different than what was happening with the regular compute cycle,” said Microsoft’s Nadella of DeepSeek’s efficiency. “We ourselves have been seeing significant efficiency gains in both training and inference for years now.” 

Nadella admitted that DeepSeek, whose model rivals the capabilities of Microsoft-backed OpenAI, “has had some real innovations” that he said should benefit his firm. “When token prices fall, inference computing prices fall. That means people can consume more, and there’ll be more apps written” that are supported by Microsoft’s infrastructure and services. 

Zuckerberg echoed Nadella’s praise for DeepSeek, saying he and his team were still digesting “a number of novel things that they did” that “we will hope to implement in our systems.”

Zuckerberg added that DeepSeek’s success vindicated Meta’s decision to make its large language model, Llama, open-source. DeepSeek was evidence, he said, that “there’s going to be an open source standard globally. And I think for our kind of own national advantage, it’s important that it’s an American standard.”

In another sign of nearly unflappable confidence in the future of America’s AI leaders, Japanese tech investor Softbank is reportedly in talks to invest up to $25 billion in OpenAI, which would make it the ChatGPT-maker’s largest investor. That would reportedly be on top of the $15 billion SoftBank already committed to spend on a joint venture with OpenAI and Oracle (ORCL), announced last week.

Update—January 30, 2025: This article has been updated with information regarding SoftBank’s potential investment in OpenAI.