Oracle Corporation’s aggressive bet on artificial intelligence and cloud infrastructure just ran into its first serious market backlash.
On Thursday, December 11, 2025, Oracle stock (NYSE: ORCL) plunged around 13–14% in U.S. trading, erasing more than $100 billion in market value in a single session and marking its worst one‑day drop since the early 2000s. [1]
The selloff came after the company’s latest fiscal second‑quarter results showed booming AI‑driven cloud demand — but also record capital spending, negative free cash flow and guidance that fell short of sky‑high expectations. Within hours, a wave of Wall Street banks cut their Oracle price targets, including Citi, Cantor Fitzgerald, Bernstein, BofA, Goldman Sachs, JPMorgan, RBC and many more. [2]
Below is a deep dive into what happened, how analysts reacted on December 11, and what it could mean next for ORCL stock and the broader AI trade.
Oracle’s latest quarter: strong AI cloud growth, harsher financial reality
Oracle’s results underscored a central tension in the AI infrastructure boom: revenue is growing fast, but the bill for building that future is arriving even faster.
Headline numbers
- Earnings beat – with a big asterisk
Adjusted earnings per share came in around $2.26, smashing Wall Street expectations of roughly $1.64. But much of that upside was driven by a $2.7 billion gain from selling Oracle’s stake in chip designer Ampere, rather than core operations. [3] - Revenue growth, but a miss
Quarterly revenue rose about 14% to ~$16.06 billion, slightly below consensus estimates near $16.1–$16.2 billion, and management’s Q3 revenue growth forecast of 16–18% fell short of what many AI‑optimistic investors had penciled in. [4] - AI and cloud are the growth engine
Oracle Cloud Infrastructure (OCI) revenue jumped 68% year over year to roughly $4.1 billion, in line with expectations, and cloud services now contribute close to half of total company revenue. However, these cloud businesses carry slimmer margins than Oracle’s legacy software, dragging overall operating margin down from about 43.4% to 41.9%. [5]
The number that scared investors: capex and free cash flow
Where Wall Street really flinched was cash going out the door:
- Record quarterly capex: Oracle spent around $12 billion on capital expenditures in the quarter — a record — largely to build out data centers and GPU‑rich infrastructure for AI workloads. [6]
- Full‑year capex explosion: Management now expects about $50 billion in capex for fiscal 2026, up from roughly $35 billion guided just three months ago — a $15 billion upward revision that stunned the market. [7]
- Negative free cash flow: The company burned about $10 billion in cash this quarter, marking its third straight quarter of negative free cash flow, as capex overwhelmed operating cash generation. [8]
Analysts and investors also zeroed in on Oracle’s heavy leverage. Various estimates put the company’s debt load at well over $110 billion, with Economic Times citing about $127 billion owed, while QuiverQuant notes roughly $111.6 billion in total debt, much of it linked to AI and cloud expansion. [9]
Massive backlog — but some disappointment
Oracle touted $523 billion in future cloud contract commitments, boosted by a huge, widely reported multi‑year AI infrastructure deal with OpenAI and other hyperscale customers. [10]
But that backlog figure actually came in a bit below what some analysts had expected, fueling worries that the AI‑driven demand curve may be flattening at the margin just as capex is going vertical. [11]
Stock reaction: worst Oracle selloff since the early 2000s
The market’s verdict was brutal:
- ORCL dropped around 13–14% on the day, trading near the low‑$190s and erasing roughly $105 billion in market value in a single session. [12]
- Barron’s noted it was Oracle’s biggest one‑day percentage decline since 2002, despite the headline EPS beat. [13]
- The steep AI‑spending hit also spilled over into the broader market: the Nasdaq and S&P 500 slipped even as the Dow hit a record, with AI‑tied names like Nvidia and AMD down 2–4%. [14]
Economic Times framed the move as an “Oracle stock crash,” highlighting the $105 billion market‑cap wipeout and warning that concerns about an AI bubble — once theoretical — have now been “revived” by Oracle’s combination of higher capex, rising debt and softer‑than‑hoped guidance. [15]
Citi, Cantor Fitzgerald and Bernstein: how the big calls shifted
The links you provided focus on three key research houses — Citi, Cantor Fitzgerald and Bernstein — which together tell the story of how sentiment evolved from pre‑earnings optimism to post‑earnings caution.
Citi: still bullish, but target cut again
Citi has been firmly in the bull camp on Oracle’s AI story, but even they are trimming expectations:
- Early December: Citi analyst Tyler Radke had already cut his price target from $415 to $375 ahead of earnings, while maintaining a Buy rating and expecting strong bookings. [16]
- December 11: Following the Q2 miss and spending surge, Radke reduced the target again to $370 from $375, reiterating a Buy. In his note, he described the quarter as a miss with “mixed” near‑term trends but emphasized that Oracle is not facing a demand problem — growth in cloud and AI workloads is still accelerating, even if profitability is under pressure. [17]
Citi’s stance in a sentence: demand looks solid, but the path to monetizing this AI super‑cycle just got bumpier and more expensive.
Cantor Fitzgerald: Overweight, but much lower AI‑adjusted upside
Cantor Fitzgerald also remains positive on Oracle — but the firm drastically reset expectations:
- On December 11, Cantor slashed its Oracle price target to $320 from $400, while keeping an Overweight rating. [18]
- The firm flagged several concerns:
- Rising AI infrastructure spending and the risk of front‑loading too much capital before cash flows catch up.
- Higher financing costs, with Oracle leaning more heavily on debt markets to fund its data‑center build‑out.
- Investor fatigue around AI infrastructure plays, now that Oracle’s negative free cash flow and heavier leverage are front and center. [19]
Cantor’s message: the long‑term AI story is intact, but the risk profile is higher and the timeline to payback is longer than bulls hoped.
Bernstein: from loud “Buy” at $364 to trimmed $339 target
Bernstein’s coverage evolved in two distinct steps:
- December 8: Bernstein reiterated an Outperform/Buy rating with a $364 price target, arguing that Oracle’s incremental business and AI cloud momentum set it up well heading into this quarter and beyond. [20]
- December 11: After the earnings release and stock collapse, MarketScreener’s analyst recap shows Bernstein cutting its target to $339 from $364 while maintaining an Outperform stance. [21]
Bernstein still sees Oracle as a structural winner in AI infrastructure — but now factors in lower near‑term margins, heavier capex and higher financing risk.
December 11: the day Wall Street rewrote its Oracle models
The most striking thing about December 11 is how many top‑tier firms moved at once.
According to MarketScreener’s consolidated analyst log for ORCL, at least 19 major banks and brokerages cut their Oracle price targets this morning alone, nearly all in response to the Q2 earnings miss, the $50 billion capex plan and the stock’s collapse. [22]
Here’s a snapshot of some of the key moves (all on December 11, 2025):
- Big U.S. banks & global houses
- AI‑enthusiast, growth‑oriented firms
- More traditional, risk‑aware shops
- Others adjusting but staying positive
What’s notable is that most of these firms kept some flavor of “Buy,” “Overweight” or “Outperform” despite lowering targets, underscoring a common thesis:
The AI and cloud demand story still looks powerful, but the risk, capital intensity and time horizon have been re‑rated.
Only a handful — notably Goldman, JPMorgan, RBC and DA Davidson — sit firmly in the Neutral/Sector Perform camp after today’s carnage. [40]
The bulls that haven’t blinked
Even after a 14% crash, Oracle still has some loud fans.
- Guggenheim: Investing.com reports that Guggenheim reiterated Oracle as its “Best Idea” with a $400 price target, framing the stock’s plunge as an overreaction to near‑term free‑cash‑flow pressure in what they see as a multi‑year AI infrastructure super‑cycle. [41]
- Jefferies and other pre‑crash bulls: Recent data compiled by QuiverQuant shows Jefferies holding a $400 target and several other banks previously in the high‑$300s before the reset; some of those will likely revisit their models in the coming days, but not all have rushed to cut yet. [42]
- Average Street view still well above spot price: TipRanks notes that, even before today’s revisions fully filter in, the average 12‑month price target sat around $346, with a high of $400 and a low of $175, based on 35 analysts — far above Thursday’s sub‑$200 trading level. [43]
In other words, the Street has sharply lowered the bar — but the bar is still set well above where the stock now trades.
Why Oracle’s AI strategy is under scrutiny
Several themes run through today’s research notes and market commentary:
1. AI infrastructure is capital‑intensive — and the bill is due now
Commentary from Barron’s, IBD and others converges on the same math: Oracle is pouring tens of billions of dollars into data centers, GPUs and networking to serve AI workloads from customers like OpenAI, Meta, Nvidia and others. [44]
That spending is generating fantastic backlog and cloud growth, but it is simultaneously:
- Crushing near‑term free cash flow (negative $10 billion this quarter),
- Pushing capex to $50 billion for the year, and
- Forcing the company to rely more heavily on debt in a world where investors have become more sensitive to leverage again. [45]
2. Backlog quality and customer concentration worries
Oracle’s $523 billion backlog is enormous, but it’s also heavily influenced by a huge, multi‑year AI contract with OpenAI. [46]
Analysts and social‑media chatter tracked by QuiverQuant point out several issues: [47]
- Customer concentration risk: If a few mega‑customers slow spending or re‑route workloads to competing clouds, that backlog could prove slower to monetize than bulls expect.
- OpenAI’s own financial uncertainty: As one Barron’s summary noted, Oracle’s backlog has been inflated by OpenAI at the same time investors are trying to handicap OpenAI’s long‑term funding needs and competitive pressures. [48]
- Inexact timing: The street has fewer doubts that AI workloads will grow; the question is when those commitments turn into high‑margin revenue and free cash flow.
3. Valuation and “AI bubble” fears
Economic Times points out that Oracle now trades at a forward P/E that’s slightly richer than Microsoft and Amazon, despite weaker balance‑sheet metrics and more volatile free cash flow. [49]
Couple that with:
- A 40%+ drawdown from Oracle’s September peak, [50]
- Talk of an “AI bubble” reignited by this week’s selloff, [51]
- And heavy dependence on execution of mega‑scale projects like Oracle’s “Stargate” data‑center initiative, [52]
…and you get a market suddenly more skeptical about paying a premium multiple for an AI infrastructure name that is spending like a hyperscaler but doesn’t yet have hyperscaler‑level profitability or cash generation.
Where consensus stands after the reset
Even after dozens of target cuts:
- Most ratings remain positive: Data from TipRanks and QuiverQuant still show the majority of covering analysts rating Oracle as some flavor of Buy/Overweight/Outperform, with only a handful in Hold/Neutral territory and very few outright Sells. [53]
- Targets are now clustered lower, but still above spot:
- High end: around $400 (Guggenheim, Jefferies). [54]
- Middle of the range: many banks now sit between $270 and $350 (BofA, TD Cowen, Stifel, Evercore, BMO, Piper, Baird, KeyBanc, Barclays, UBS). [55]
- Low end: DA Davidson at $180, Goldman at $220, JPMorgan at $230, RBC at $250, all Neutral/Sector Perform. [56]
The result: consensus has shifted from “explosive AI winner” to “high‑beta AI infrastructure play with real execution risk” — but not to “broken story.”
Key questions for Oracle (and ORCL stock) going forward
For investors and readers watching Oracle as a bellwether for AI infrastructure, the next few quarters will hinge on a handful of datapoints:
- Capex trajectory vs. revenue growth
Does Oracle stick with a $50 billion capex plan, or does management trim spending if revenue and backlog conversion don’t accelerate in tandem? - Free cash flow inflection
The Street will be looking for clear visibility on when free cash flow turns sustainably positive again — and how quickly it can scale back toward pre‑AI levels. - Backlog conversion and customer diversification
How quickly does the $523 billion backlog translate into recognized revenue, and does Oracle reduce reliance on any single mega‑customer such as OpenAI? - Margin stabilization
OCI growth is impressive, but margins are thinner than in traditional software. Investors will want to see whether Oracle can expand gross and operating margins even while it absorbs higher data‑center costs. [57] - Competitive dynamics
How effectively can Oracle compete with hyperscalers like Microsoft, Amazon and Google Cloud, especially as those players deepen their own AI and database offerings? - Leadership and communication
This is one of the first truly major market tests for Oracle’s new co‑CEO structure, with Clay Magouyrk and Mike Sicilia sharing the top job and Safra Catz shifting to Executive Vice Chair. Their ability to clearly articulate the AI investment roadmap — and deliver against it — will be critical. [58]
Bottom line: Oracle is still all‑in on AI — the market just repriced the risk
Oracle’s December 11 crash isn’t about a collapse in AI demand. Cloud and AI workloads are growing rapidly, and Oracle’s backlog and OCI numbers underline that reality.
What changed is the market’s tolerance for the cost and risk of building the infrastructure behind that boom:
- Capex and debt are higher than most investors expected.
- Guidance was softer than an AI‑hype‑driven market had priced in.
- And virtually every major Wall Street bank just admitted, via lower price targets, that their previous assumptions were too optimistic.
For long‑term followers of AI infrastructure, Oracle now looks like a high‑conviction but high‑volatility test case: if its $50 billion bet on AI data centers and hyperscale cloud pays off, today’s price reset may eventually look like an opportunity. If not, it will stand as a stark warning about the dangers of chasing AI headlines without watching the cash‑flow math.
References
1. www.barrons.com, 2. www.marketscreener.com, 3. www.barrons.com, 4. www.barrons.com, 5. finance.yahoo.com, 6. www.barrons.com, 7. markets.financialcontent.com, 8. www.investors.com, 9. m.economictimes.com, 10. www.barrons.com, 11. m.economictimes.com, 12. m.economictimes.com, 13. www.barrons.com, 14. www.investopedia.com, 15. m.economictimes.com, 16. www.quiverquant.com, 17. www.tipranks.com, 18. www.investing.com, 19. www.investors.com, 20. www.insidermonkey.com, 21. www.marketscreener.com, 22. www.marketscreener.com, 23. www.marketscreener.com, 24. www.marketscreener.com, 25. www.marketscreener.com, 26. www.tipranks.com, 27. in.investing.com, 28. www.marketscreener.com, 29. www.marketscreener.com, 30. www.marketscreener.com, 31. www.marketscreener.com, 32. www.tipranks.com, 33. www.investing.com, 34. www.investing.com, 35. www.marketscreener.com, 36. www.investing.com, 37. www.tipranks.com, 38. www.marketscreener.com, 39. www.marketscreener.com, 40. www.marketscreener.com, 41. www.investing.com, 42. www.quiverquant.com, 43. www.tipranks.com, 44. www.barrons.com, 45. www.investors.com, 46. www.barrons.com, 47. www.quiverquant.com, 48. www.barrons.com, 49. m.economictimes.com, 50. www.investors.com, 51. www.marketscreener.com, 52. www.barrons.com, 53. www.tipranks.com, 54. www.investing.com, 55. www.marketscreener.com, 56. www.marketscreener.com, 57. www.barrons.com, 58. markets.financialcontent.com
