Oracle’s stock tumbled nearly 11% in premarket trading on Thursday as a wave of investor caution hit the company. The real trigger? lower-than-expected forecasts, a fresh jump in capital spending, and rising concerns that Oracle’s massive bet on artificial intelligence could take much longer than expected to generate returns.
The company shocked markets by warning that capital expenditures for fiscal 2026 would be $15 billion higher than the $35 billion it had estimated just months earlier. The new number pushes expected spending to about $50 billion, largely tied to building giant AI data centers across the US
Is Oracle taking on too much financial risk? Is the AI boom inflating into a bubble? And can the company convert its huge pipeline of AI contracts into real revenue without overburdening its balance sheet?
According to Morningstar, “We have mixed feelings about the scale of data center investments Oracle has planned over the coming years. The analysts added that if enthusiasm for AI cools and “key customers like OpenAI reduce their computing demand,” Oracle could struggle to find other workloads to fill the gap.
Oracle’s quarterly results
The company’s quarterly revenue of $16.06 billion, slightly below analysts’ expectations of $16.21 billion. The miss was small, but in a market priced for perfection, it has shaken investor’s confidence, Capital expenditures jumped to $12 billion, far above Wall Street’s forecast, and the company lifted future spending plans sharply. On the earnings call, management stressed that Oracle must invest at scale to deliver on its AI contracts, even if that hurts margins in the short term.
As eToro’s Farhan Badamin told Reuters, “This will be a question of patience for investors. This AI boom won’t be an overnight success, and spending in the short term is a necessity, but it will pressure margins.”
Investors were already nervous after Oracle raised $18 billion in a jumbo bond sale in September, one of the biggest tech-sector debt offerings ever and secured billions more through construction loans for US data centers.
A huge AI pipeline
On one key metric, Oracle actually exceeded expectations, the company reported $523 billion in future cloud contracts, beating consensus and marking a 438% increase from last year. New contracts from Meta and Nvidia helped diversify its base beyond OpenAI. But turning those contracts into revenue requires building expensive AI infrastructure.
CEO Clay Magouyrk stated new models that reduce upfront costs, “One of them is that customers can actually bring their own chips, and in those models, Oracle obviously doesn’t have to incur any capital expenditures upfront for that model.” CFO Doug Kehring also tried to calm nerves, saying Oracle has multiple funding options and “borrow substantially less than most people are modeling.” He reiterated the company’s pledge to maintain an investment-grade rating.
Oracle’s five-year, $300 billion cloud deal with OpenAI helped push the stock to record highs in September. For a moment, Larry Ellison even became the world’s richest man. But the same partnership is now raising concerns about overexposure, especially as OpenAI faces questions about profitability and governance.
“It’s telling us that many of these tech stocks are priced for absolute perfection, and when there’s a revenue miss, which was a small revenue miss to be fair by Oracle yesterday – they do get penalised,” Wealthify consultant Colleen McHugh told Reuters.
Even Larry Ellison cautioned about the future of AI technology itself, “There are going to be a lot of changes in AI technology over the next few years and we must remain agile in response to those changes.”
He also promoted a strategy of “chip neutrality,” saying Oracle will buy whatever chips customers want. “We will continue to buy the latest GPUs from Nvidia, but we need to be prepared and able to deploy whatever chips our customers want to buy.”