Oracle stock on pace for worst quarter since 2001, AI concerns

Three months ago Oracle named Clay Magouyrk and Mike Sicilia as its new CEOs. They’re off to a rough start.

Oracle shares have plummeted 30% so far this quarter. With four trading days remaining in the period, the stock is on pace for its sharpest decline since 2001 and the dot-com bust.

Investors have grown skeptical about the database software vendor’s ability to open more server farms for ChatGPT operator OpenAI, which agreed in September to spend over $300 billion with Oracle.

Earlier this month, Oracle reported weaker-than-expected quarterly revenue and free cash flow. On the earnings call, newly appointed finance leader Doug Kehring called for $50 billion in fiscal 2026 capital expenditures, 43% higher than the plan in September and double the total from a year earlier. Additionally, Oracle is plotting $248 billion in leases to boost cloud capacity, on top of building data centers.

Such growth will require boatloads of debt. In September, Oracle raised $18 billion in a jumbo bond sale, one of the largest debt issuances on record in the tech industry. Kehring committed on the earnings call to keeping Oracle’s investment-grade debt rating. But some skeptical investors are betting otherwise, pushing up the prices of Oracle’s credit default swaps.

“Considering Oracle is already barely hanging on to an investment grade rating, we would be concerned about Oracle’s ability to live up to these obligations without restructuring its OpenAI contract,” analysts at DA Davidson wrote in a note to clients on Dec. 12. They have the equivalent of a hold rating on the stock.

Oracle declined to comment.

Magouyrk and Sicilia’s tenure began at a time of historic optimism.

About two weeks before they took the reins from Safra Catz, Oracle reported a 359% revenue backlog tied heavily to OpenAI’s commitment. The deal that represented a major endorsement for Oracle, which was left off Gartner’s list of top five cloud infrastructure providers by revenue for 2024.

Following reports about the OpenAI agreement on Sept. 10, Oracle’s stock shot up almost 36%, the third-sharpest rally since the company’s 1986 IPO. Shares reached an intraday record of $345.72.

“We think $340 was terrifying,” said Zachary Lountzis, vice president at Lountzis Asset Management, in an interview. Lountzis held $25 million in Oracle shares as of Sept. 30, according to a filing.

The stock has since lost 43% of its value, closing on Wednesday at $197.49, though it got a bump last Friday after TikTok said it had agreed to sell part of its U.S. business to Oracle and other investors. Oracle has for years delivered cloud services to TikTok.

Lountzis said his team first bought Oracle shares in 2020, when the stock was below $60. It’s held onto it stake through the recent highs and lows, picking up another roughly 30,000 shares in the first quarter of this year.

“Our philosophy is that we’re OK with short-term over-valuation if the economics of the business have not changed, and that was the case with Oracle,” Lountzis said. “We didn’t feel the economics of the business changed with all the largely positive news that came out. And I think what we’ve seen from $340 down to $180 is actually a very healthy correction.” 

For Lountzis, much of his trust in the company comes down to Larry Ellison, who founded Oracle in 1977 and is now the world’s second-richest person, according to Bloomberg.

“You would have gone bankrupt 40 times betting against Larry over the last 50 years,” Lountzis said. “He sees the future.”

In October, Sicilia, Magouyrk and Kehring laid out a vision for a much faster-growing Oracle, with revenue set to step up to $225 billion in the 2030 fiscal year from $57 billion in fiscal 2025. Most of that growth will come from AI infrastructure, with Nvidia’s graphics processing units at the center of it.

But while Magouryk was telling analysts to prepare for “hypergrowth,” such expansion would come at the expense of profitability, because Oracle’s core software business commands much higher margins.

In fiscal 2021, Oracle’s gross margin was 77%. Analysts polled by FactSet see it falling to about 49% in 2030, with about $34 billion in total negative free cash flow over the next five years before that figure turns positive in 2029.

Eric Lynch, managing director at Florida’s Suncoast Equity Management, said it’s hard as an investor to get comfortable with Oracle’s plans.

“Four or five years is a long time,” Lynch said. “That’s just not within our investment discipline.”

Lynch also said he’s worried about such heavy dependance on OpenAI, which is burning cash at a rapid rate and has committed to over $1.4 trillion in total AI buildouts and investments.

“Will the demand be there from OpenAI?” Lynch said.

Wells Fargo analyst Michael Turrin launched coverage of Oracle earlier this month with the equivalent of a buy rating and a $280 price target. He said the industry’s perception will likely improve if Oracle follows through with OpenAI, which could could account for over one-third of the company’s revenue by 2029, according to Turrin’s estimate.

“They’re kind of shifting away from more of a value-oriented business to a more growth-oriented business,” Turrin said.

A big challenge for Oracle remains picking up market share in cloud infrastructure, where the company badly trails Amazon, Microsoft and Google even though its customer roster includes names like Meta, Uber and Elon Musk’s xAI.

Databricks, which was just valued at $134 billion in a funding round, doesn’t make its popular data processing software available on Oracle’s cloud.

That will happen “when customers start banging on my door, saying, ‘You need to run on Oracle,'” Databricks CEO Ali Ghodsi said in an interview. “Maybe it’s getting there, but we just haven’t heard that.”

Databricks rival Snowflake hasn’t brought its services to Oracle either.

Turrin said that Oracle’s credibility in the market will hinge on the success of its AI buildout.

“Then customers start to look at this and say, wow, this company was trusted to build some of the largest training clusters in the world, and they’re delivering on them,” Turrin said. “We should take a look at that too and figure out what’s happening here.”

WATCH: How the massive power draw of generative AI is overtaxing our grid