Updated: Friday, Dec. 12, 2025 (midday U.S. session). Prices cited below reflect quotes around 12:25 p.m. ET and may move quickly.
Wall Street’s AI trade is getting stress-tested again on Friday — and the pressure point isn’t demand for AI, but what investors are willing to pay for it.
By midday, AI-chip and AI-infrastructure leaders are leading the downside after Broadcom’s margin warning and renewed unease around Oracle’s debt-funded AI buildout, prompting a broad pullback in the Nasdaq-heavy growth complex. At the same time, the broader market remains within reach of record territory, with investors rotating into sectors perceived as less exposed to AI valuation risk. [1]
Midday market snapshot: AI and semiconductors are underperforming
At around midday Friday, tech and semiconductors are lagging the broader tape:
- S&P 500 proxy (SPY): down about 0.96%
- Nasdaq-100 proxy (QQQ): down about 1.67%
- Semiconductor ETFs (SOXX, SMH): down roughly 3.5%–3.7%
That “semis first” weakness matters for the AI narrative because chips and data-center buildouts have been the most direct, most crowded expression of AI optimism in public markets.
AI stock prices today: key names and what’s moving them
Below is a quick look at major U.S.-listed AI bellwethers and AI-adjacent names around midday (quotes near 12:25 p.m. ET):
AI chips and infrastructure
- Broadcom (AVGO): down about 10.3%
- Nvidia (NVDA): down about 2.2%
- AMD (AMD): down about 4.0%
- Arm (ARM): down about 2.9%
Cloud and Big Tech “AI platforms”
- Oracle (ORCL): down about 5.0%
- Microsoft (MSFT): down about 1.0%
- Alphabet (GOOGL): down about 1.1%
- Amazon (AMZN): down about 1.6%
- Meta (META): down about 1.5%
AI software and “AI momentum” names
- Palantir (PLTR): down about 2.7%
- C3.ai (AI): down about 3.9%
- SoundHound AI (SOUN): down about 1.8%
The takeaway is straightforward: this is a broad de-risking move inside the AI complex, not a single-stock story — even though the catalysts are very stock-specific.
Why AI stocks are selling off today
1) Broadcom’s warning: AI growth, thinner margins
Broadcom is being treated as a sentiment barometer for “AI profit pools,” and Friday’s drop reflects a classic market message: growth is good; margins still matter.
Reuters reported that Broadcom projected revenue above estimates but flagged pressure on gross margin as AI becomes a larger mix of sales, with management pointing to a sequential margin dip of about 100 basis points. [2]
The concern isn’t that AI demand is disappearing — it’s that custom AI silicon and AI system sales may be less lucrative (or at least less cleanly accretive) than the market assumed at peak enthusiasm. [3]
A counterpoint from analysts: Morningstar argued the margin dilution narrative may be overdone and raised its fair value estimate on Broadcom to $480 per share (from $365) on stronger AI-chip expectations. [4]
2) Oracle: the market is questioning the “debt-fueled AI hyperscaler” bid
Oracle’s latest wave of volatility continues to ripple through AI stocks because it sits at a sensitive junction: cloud infrastructure + OpenAI-linked demand + very large capex.
On Friday, Reuters reported that Oracle has pushed back completion dates for some OpenAI-related data centers to 2028, citing a Bloomberg report, and noted the delay was attributed to labor and material shortages. [5]
More importantly for equity investors, Reuters highlighted that Oracle recently warned fiscal 2026 capex would be $15 billion higher than it estimated just months earlier — a sharp revision that revived debate over how quickly AI infrastructure spending turns into durable returns. [6]
And in a Breakingviews column, Reuters underscored the cash strain dynamic: Oracle’s capital expenditure vs. operating cash flow imbalance is now a central part of the bear case as investors re-check the math behind “AI at any cost.” [7]
3) Higher yields + high valuations: the classic growth-stock headwind is back
When bond yields rise, richly valued growth stocks often feel it first — and today’s tape is a reminder. The Associated Press noted the 10-year Treasury yield moved higher as stocks fell, amplifying the valuation debate around AI leaders. [8]
In other words: even if AI fundamentals remain strong, the discount rate can still bite.
“AI bubble” fears are rising — but investors aren’t treating it like a 2000-style unwind (yet)
A Reuters analysis published Friday captured the market’s split-screen view:
- Yes, investors are becoming more selective about AI names and less willing to reward capex headlinesautomatically.
- No, positioning and short interest don’t yet look like the market is “all-in” on betting the AI story collapses imminently — especially in the biggest AI beneficiaries. [9]
Reuters also pointed to the idea that broad market strength has helped cushion index-level damage even as mega-cap AI winners wobble — a sign the rally is broadening beyond the AI trade. [10]
Still, prominent skeptics are getting airtime again. Business Insider reported comments from investor Michael Burrywarning that AI bubble timing is unknowable and that shorting can be dangerous even if valuations look extreme. [11]
The hidden catalyst behind today’s jitters: AI infrastructure is turning into a debt story
One of the most underappreciated developments in the AI boom is how quickly it’s reshaping credit markets.
Reuters flagged that AI data-center financing has surged — citing a UBS report that pegs the jump at $15 billion in 2024 to $125 billion in 2025, with more supply expected into 2026. [12]
The same Reuters piece highlights a widening toolkit of funding methods — from investment-grade issuance by tech firms to private credit and asset-backed securities linked to data-center rents. [13]
For equity investors, this matters because it reframes a key question:
Is the AI buildout a self-funding flywheel — or a leverage-dependent race that becomes fragile if capital costs rise?
Oracle is simply the most visible flashpoint right now — but the market is clearly beginning to separate AI “haves”(cash-rich hyperscalers) from AI “have-nots” (players that must borrow heavily to keep up). [14]
Two big cross-currents investors are watching today: regulation and geopolitics
A) Washington moves toward a single federal AI framework
Policy is back in focus after President Donald Trump signed an executive order aimed at curbing state-by-state AI regulation, with the White House signaling it wants a single national framework and plans to work with Congress on legislation. [15]
The market relevance is nuanced:
- A unified framework could reduce compliance friction for Big Tech AI rollouts.
- But the approach is politically contentious and could face pushback from states and advocacy groups, injecting legal and regulatory uncertainty. [16]
B) Nvidia and China demand: bullish headline, complicated reality
Nvidia had its own headline catalyst: Reuters reported Nvidia told Chinese clients it is evaluating adding H200 production capacity after demand exceeded current output, following a U.S. decision to allow H200 exports to China with a 25% fee. [17]
Even if demand is strong, the story also highlights ongoing constraints: supply chain capacity, export policy risk, and the geopolitical push-pull that can quickly reprice semiconductor expectations.
The product race isn’t slowing: OpenAI’s latest model adds to the competitive pressure on Big Tech
While public markets debate ROI and margins, the AI product cycle keeps accelerating.
Reuters reported OpenAI launched GPT-5.2, positioning it as a step forward in coding and long-context tasks, amid intensified competition with Google’s Gemini. Reuters also noted Disney’s $1 billion investment in OpenAI tied to content licensing for its video-generation tools. [18]
For stocks, the implications ripple outward:
- Microsoft remains tightly linked to OpenAI’s ecosystem.
- Alphabet is both a model competitor and a major AI platform in its own right.
- Media and IP owners are increasingly exploring paid partnerships — a potential monetization path the market wants to see more of. [19]
Not all “AI” stocks are down: Rivian’s AI Day shows the theme still sells — selectively
Even on a down day for AI megacaps, one pocket of the market surged: Investors.com reported Rivian jumped after announcing an in-house AI inference chip and a next-gen autonomy roadmap at its “Autonomy & AI Day.” [20]
This is a useful reminder of where the market is in late 2025: the AI theme still commands attention — but investors are drawing a sharper line between credible execution stories and capex-without-clarity stories.
What to watch into the close and into next week
For investors tracking AI stocks and the broader U.S. market, the next catalysts are less about “Is AI real?” and more about “Who earns the best returns on AI?”
Key items to watch:
- Broadcom follow-through: Will dip buyers accept margin dilution if backlog and AI revenue growth remain strong? [21]
- Oracle’s financing narrative: Any additional detail on OpenAI-linked buildouts, timelines, and funding could move both the stock and AI sentiment broadly. [22]
- Nvidia supply and policy: China demand headlines are supportive, but export terms and supply constraints can turn quickly. [23]
- Rates and rotation: If yields stay elevated, the market may continue rotating away from the most expensive AI trades toward “value” and cyclicals. [24]
- Regulatory headlines: The White House push for a single AI framework could become a meaningful overhang — or a tailwind — depending on legal and congressional trajectory. [25]
Bottom line
At midday on Dec. 12, 2025, the U.S. AI stock selloff looks less like a demand shock and more like a profitability and financing reality check: margins, debt, capex discipline, and time-to-payoff are back at the center of the AI narrative. [26]
References
1. www.reuters.com, 2. www.reuters.com, 3. www.reuters.com, 4. www.morningstar.com, 5. www.reuters.com, 6. www.reuters.com, 7. www.reuters.com, 8. apnews.com, 9. www.reuters.com, 10. www.reuters.com, 11. www.businessinsider.com, 12. www.reuters.com, 13. www.reuters.com, 14. www.reuters.com, 15. www.reuters.com, 16. apnews.com, 17. www.reuters.com, 18. www.reuters.com, 19. www.reuters.com, 20. www.investors.com, 21. www.reuters.com, 22. www.reuters.com, 23. www.reuters.com, 24. apnews.com, 25. www.reuters.com, 26. www.reuters.com
