By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel is largely bearish on Oracle's hyperscaler pivot due to high debt, negative FCF, and questionable execution. The $553B backlog is seen as fragile, and there are concerns about Oracle's ability to compete with better-capitalized hyperscalers and maintain high margins.
Risk: High debt service and potential revenue misses
Opportunity: Securing long-term government contracts
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
To put it mildly, 2026 has not been a good year for Big Software so far. Even the giants in the industry have taken serious hits to their stock prices. One stark example of this is Oracle's (NYSE: ORCL) near-21% year-to-date decline.
The rout is due largely to investor fears that tech companies identified with legacy solutions -- like Oracle, with its databases -- will be swept up in a great wave of disruption from artificial intelligence (AI) models that can do their work better, cheaper, and quicker. Yet Oracle is actually pivoting toward an AI-heavy business model. So does the sell-off make the best bargain AI stock these days?
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Oracle of the future
Oracle's big gamble is that it can transition from its traditional wheelhouse of database and software-as-a-service (SaaS) solutions into a massive AI infrastructure hyperscaler (i.e., operator of large-scale data centers). In short, it aims to be a major -- if not the major -- landlord in the AI revolution.
Speaking of AI, Oracle also wants to become a prominent vendor in the space. It envisions developing industry-specific AI "agents," i.e., next-generation models that can autonomously perform tasks for the client. The company also aims to become a purveyor of "sovereign AI" solutions for national governments that, instead of being housed on company servers, would rest entirely within the client's IT assets.
That next-generation technology should help Oracle trim costs, too. It has shrunk its developer teams through a series of layoffs, as it can replace humans with AI coding tools to handle much of the grunt work.
The financials behind this tectonic shift are, well, Oracle-sized. Recently, management drastically upped its estimate for its current fiscal year capital expenditures (capex) to $50 billion. For perspective, keeping the company's legacy database business alive and humming used to cost it less than $2 billion annually.
Even with the company's mountains of revenue, it will need some help reaching its suddenly more ambitious goals. In February, it announced a package of financing moves to raise greenbacks for this initiative. It issued $25 billion in debt, matched by $25 billion in secondary share issues and convertible stock.
The big ramp-up in spending isn't doing wonders for free cash flow (FCF). Very uncharacteristically for this historically cash-generating monster, Oracle's FCF was deep in the red, to the tune of almost $44 billion through the first three quarters of its current fiscal 2026.
At the end of its most recently reported (third) quarter, long-term debt had ballooned to almost $148 billion, almost 50% higher than the end-2025 tally of under $100 billion. It's also nearly double the same metric at the conclusion of fiscal 2022, which is hardly a lifetime ago.
Stargate star
To be sure, there's plenty of demand for huge AI data centers, and Oracle's legacy business isn't (yet) being affected by that feared AI disruption. For the entirety of fiscal 2026, it's expecting revenue of $67 billion, comfortably above the nearly $57.4 billion it raked in last year. That's anticipated to leap to $90 billion in fiscal 2027 as more of those facilities come onstream.
To be sure, Oracle has customers eager to do business with it as a hyperscaler landlord. It's the AI facility provider of choice for the Stargate Project, a federal government-supported initiative to bolster U.S. leadership in AI.
One foundational deal within the project is Oracle's commitment to build out significant AI data center capacity for leading developer OpenAI, of ChatGPT fame. Oracle surprised investors in that recently reported quarter with a 325% year-over-year surge in project backlog ("remaining performance obligations," in the company's words) to a huge $553 billion; its OpenAI work is a major part of this.
So is Oracle a buy?
For years, despite its size, power, and prominence, Oracle wasn't particularly exciting as an investment. It was a reliably steady and profitable performer.
That was then, however, and this is now. The AI revolution is still in its early stages, and Oracle's play is to get a jump on the still-massive investments into infrastructure needed to advance it. Will it be dominant in a world of other data center operators and big, wealthy tech companies who might want to compete for the same prize?
Personally, I'd give it a better-than-average chance since first-mover advantage can be the deciding factor in the tech world. After all these years, company founder Larry Ellison still has a huge stake in the company, so he and his team don't really need to convince others of the plan's viability; they can simply implement it.
In short, Oracle is one of the biggest gambles in the tech world these days (and, to an extent, the entirety of corporate America). It's taken big leaps before, which is why it's such a long-time powerhouse in databases.
Given that, I'd place a bet on Oracle stock, yet I have to emphasize that the future is still up in the air. Because of that, despite my bullishness, I wouldn't say Oracle is the most undervalued AI play on the scene just now.
Should you buy stock in Oracle right now?
Before you buy stock in Oracle, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Oracle wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $494,747!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,094,668!*
Now, it’s worth noting Stock Advisor’s total average return is 911% — a market-crushing outperformance compared to 186% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
Eric Volkman has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Oracle. The Motley Fool has a disclosure policy.
Four leading AI models discuss this article
"Oracle's hyperscaler transition requires it to sustain massive capex while proving it can compete on unit economics against entrenched cloud giants—a bet the market is right to discount given the cash burn trajectory and unproven margin profile."
Oracle's $50B capex pivot is real and the $553B backlog (driven by OpenAI/Stargate) is material. But the article buries the knife: -$44B FCF through Q3 on a $67B revenue run-rate means Oracle is burning cash at 66% of sales. Debt nearly doubled to $148B in one year. The legacy database business (still ~40% of revenue) masks whether the hyperscaler economics actually work. The article assumes first-mover advantage in data centers, but AWS, Azure, and Google have entrenched relationships, lower capex ratios, and proven operating leverage. Oracle is betting its balance sheet on winning a hyperscaler war against better-capitalized competitors.
If Stargate scales as promised and Oracle achieves even 15% gross margins on $90B+ revenue by FY2027-28, FCF turns positive and debt becomes manageable; the stock's 21% decline may look prescient buying.
"Oracle's transition into a capital-intensive hyperscaler is fundamentally eroding its historical status as a high-margin, cash-generative software business, creating unsustainable leverage risks."
Oracle (ORCL) is attempting a high-stakes pivot from a legacy software cash-cow to a capital-intensive infrastructure utility. While the $553 billion backlog is impressive, the $44 billion FCF deficit and debt ballooning to $148 billion signal a massive execution risk. Oracle is essentially levering its balance sheet to compete with hyperscalers like Microsoft and AWS that have deeper pockets and existing cloud ecosystems. The market is rightfully punishing the stock; this isn't just 'AI disruption' fear, it's a fundamental concern regarding capital allocation and the return on invested capital (ROIC) for these massive data center build-outs in a high-interest-rate environment.
If Oracle successfully secures long-term, high-margin 'sovereign AI' contracts with governments, the massive capex will look like a masterstroke of infrastructure positioning rather than a reckless gamble.
"Oracle’s AI hyperscaler bet is plausibly transformative but is a binary, execution-dependent risk because massive capex, rising debt, lumpy backlog recognition, and potential dilution could swamp the upside if timelines or margins slip."
Oracle’s pivot from databases to become an AI hyperscaler is real and backed by large contracts (notably OpenAI) and a $553 billion backlog figure, but the story is far from straightforward. Management has increased fiscal 2026 capex guidance to ~$50 billion and funded the push with ~$25 billion debt plus equity/convertibles, sending FCF roughly -$44 billion through nine months and long-term debt to ~$148 billion. The trade-off: if Oracle can bring facilities online, grow revenue toward the company’s $90 billion FY2027 target and regain positive FCF, upside is material; if projects are delayed, margins compress, or government/enterprise sales slow, the balance sheet risk and dilution could overwhelm any re-rating.
If Oracle executes flawlessly — delivering facilities on schedule to OpenAI and other large clients while monetizing sovereign-AI contracts — revenue and FCF could swing positive quickly and justify a much higher multiple. Moreover, Ellison’s large equity stake reduces risk of value-destroying management decisions and signals alignment with shareholders.
"Oracle's ballooning $148B debt and -$44B FCF expose acute balance sheet risks that could derail its AI pivot before hyperscaler dreams materialize."
Oracle's (ORCL) AI hyperscaler bet hinges on $50B FY2026 capex (vs. prior $2B for legacy databases), fueling a $553B RPO surge including Stargate/OpenAI, with revenue eyed at $67B this year and $90B FY2027. But the article downplays red flags: FCF plunged to -$44B in 3Q FY2026, long-term debt hit $148B (up 48% from end-FY2025), after $25B debt + $25B equity raises. This leverage spike amid 5%+ rates risks distress if AI demand falters or capex overruns. Legacy databases face real AI cannibalization long-term, and ORCL trails AWS/Azure in scale. Market's 21% YTD drop prices in execution peril.
A $553B backlog (up 325% YoY) and government-backed Stargate/OpenAI deals provide multi-year revenue visibility, potentially de-risking the capex if AI infrastructure spend exceeds $1T annually as projected.
"Backlog visibility matters less than contract stickiness and margin sustainability when leverage is this high."
Everyone assumes the $553B backlog is durable, but nobody's stress-tested contract cancellation risk. OpenAI's Stargate is partially government-backed; if political winds shift or AI capex ROI disappoints, those contracts could face renegotiation or delays. Oracle's debt service at current rates (~$7-8B annually) leaves little room for revenue miss. The 'flawless execution' scenario requires not just on-time delivery but sustained demand at premium margins—a two-variable bet masquerading as one.
"Oracle's pivot to sovereign AI creates a moat that overrides traditional hyperscaler competitive dynamics and mitigates revenue volatility."
Anthropic is right to highlight the contract fragility, but Grok and Google are ignoring the 'sovereign AI' moat. These aren't standard enterprise contracts; they are utility-level infrastructure commitments often tied to national security. If Oracle pivots to being the 'government cloud' provider, their pricing power becomes inelastic, insulating them from the volatility of consumer AI demand. The debt is a weapon, not just a liability, provided they secure these sticky, multi-decade state-level dependencies.
"Sovereign/government contracts are unlikely to provide the fast, high-margin revenue needed to justify Oracle's debt-fueled capex."
Treating 'sovereign AI' as a cure-all is risky. Government contracts are often long, bureaucratic, price-capped, and carry heavy compliance/customization costs that compress margins and lengthen revenue recognition. They also incentivize multi-vendor sourcing for resilience, undermining exclusivity. Oracle's debt-fueled capex needs rapid, high-margin monetization — sovereign deals are unlikely to deliver that timing or margin profile, so leaning on them as justification for heavy leverage is misplaced.
"Sovereign AI hype ignores Oracle's weak gov't wins, multi-vendor realities, and back-loaded RPO that prolongs FCF burn."
Google's 'sovereign AI moat' overlooks Oracle's mediocre government track record—lost major CIA bid to AWS—and multi-vendor RFPs that cap exclusivity. OpenAI correctly flags bureaucracy, but nobody notes RPO recognition is back-loaded: even sticky contracts mean FCF stays negative through FY2027, with $148B debt at 5%+ rates demanding $7.5B+ annual service before capex.
The panel is largely bearish on Oracle's hyperscaler pivot due to high debt, negative FCF, and questionable execution. The $553B backlog is seen as fragile, and there are concerns about Oracle's ability to compete with better-capitalized hyperscalers and maintain high margins.
Securing long-term government contracts
High debt service and potential revenue misses