Dive Brief:
- Oracle intends to invest $35 billion to build out its cloud capacity this year, which is $10 billion more than previously planned, as it races to meet demand for its compute services, CEO Safra Catz said Tuesday, during a Q1 2026 earnings call. Most of the spending will go into revenue-generating data center equipment, the executive said.
- The company’s remaining performance obligations for unfulfilled services under contract more than quadrupled year over year to $455 billion, driven by massive cloud contracts with a “who’s who of AI,” including OpenAI, xAI, Meta and Nvidia, Catz said. The company’s cloud segment revenues grew 27% to $7.2 billion during the three months ending Aug. 31.
- “All this money we’re spending on training is going to have to be translated into products that are sold, which is all inferencing,” Oracle CTO and Chairman Larry Ellison said. “The AI inferencing market will be much, much larger than the AI training market.”
Dive Insight:
Massive spending on AI infrastructure to train large language models and process enterprise workloads are the new normal in the cloud business.
The three biggest players by market share — AWS, Microsoft and Google — upped the ante earlier this year, eclipsing Oracle with pledges to increase their capital investments to $100 billion, $80 billion and $85 billion, respectively.
Spending on data center infrastructure spiked 51% in 2024 to $455 billion. The upward trend continued through the first three months of the year, according to Dell’Oro Group research.
Oracle’s more modestly sized cloud dominion is expanding on several fronts, with large, medium and small buildouts.
The junior hyperscaler reportedly inked a five-year, $300 billion deal with OpenAI for Oracle cloud services, according to a Wall Street Journal report Wednesday. An Oracle spokesperson declined to comment on the report in a Wednesday email to CIO Dive. In July, the company teamed up with OpenAI to develop 4.5 gigawatts of data center capacity as part of the White House-endorsed, $500 billion Stargate AI initiative.
Oracle has also been busy deploying its database servers in AWS, Azure and Google Cloud facilities. The multicloud arrangement is operational in 34 regions, up from 23 the prior quarter. The company has 37 additional deployments planned, according to Catz, who said revenue from the partnerships grew 1,529% in Q1.
As enterprise AI projects ramp up, Oracle is banking on data security concerns to drive workloads into private cloud environments.
“We have large companies that are buying basically their own Oracle Cloud regions, in fact multiple Oracle Cloud regions, because they don’t want to have any neighbors in their cloud,” Ellison said. “We can give you a private version of the Oracle Cloud with every feature, every security feature, every function, everything we do for $6 million.”
While private cloud isn’t an inexpensive option, it’s a viable alternative for organizations grappling with data privacy and sovereignty hurdles, Keith Kirkpatrick, research director for enterprise software and digital workflows at Futurum Group, told CIO Dive in an email.
“It is a play to attract customers that, for security, regulatory, or other issues, simply aren’t going to be attracted to a public cloud,” Kirkpatrick said.
