Reports on Tuesday morning that OpenAI missed internal goals on user targets and revenue have rattled the AI infrastructure complex. Per Bloomberg, as flagged in a BTIG morning note, investors are questioning whether outsized data-center CapEx commitments from hyperscalers can hold. Tech futures opened lower, and the Philadelphia Semiconductor Index snapped an 18-day winning streak. The SOX itself was last quoted at 441.45, down 3% on the session.
BTIG’s Jonathan Krinsky warned on April 20 that semiconductors were “dangerously overheated,” with the SOX +40% above its 200-day moving average, the most overbought reading since June 2000. Goldman Sachs separately noted that AI disruption concerns are prompting renewed scrutiny of long-term valuations. With Microsoft and Meta reporting Wednesday after close and Amazon and Apple Thursday after close, this week is the real stress test.
Below are the five U.S.-listed stocks most directly tethered to AI infrastructure spend. This is a map of where the risk sits if the CapEx narrative wobbles.
1. NVIDIA (NASDAQ: NVDA)
NVIDIA (NASDAQ:NVDA | NVDA Price Prediction) is the most direct beneficiary and most exposed name in the group. Q4 FY2026 revenue hit $68.1 billion, up 73% YoY, with Data Center revenue of $62.3 billion and large cloud service providers accounting for just under 50% of that segment. The Q1 FY2027 guide is ~$78 billion, and CEO Jensen Huang called this “the agentic AI inflection point”.
Bull case: multi-year deals with Meta spanning millions of Blackwell and Rubin GPUs, plus a 10-gigawatt OpenAI commitment. Risk: Shares trade at a P/E of 43 after a 95% one-year run. Polymarket implies an 82% probability NVDA closes lower today. For long-term holders, the question is whether hyperscaler outlays can stay vertical.
2. Broadcom (NASDAQ: AVGO)
Broadcom (NASDAQ:AVGO) recently crossed $2 trillion in market cap on custom AI silicon momentum. Q1 FY2026 revenue was $19.31 billion, with AI semiconductor revenue of $8.40 billion, up 106% YoY. CEO Hock Tan has publicly targeted exceeding $100 billion in AI sales by 2027.
The bull case is XPU concentration with Google, Meta, and growing hyperscalers. Risk is the same concentration: a handful of customers drive the AI line. Reddit sentiment on AVGO flipped from sustained bullish readings of 70-78 down to a 25 (bearish) overnight. Shares fell 3% Tuesday morning after a 108% one-year gain.
3. Oracle (NYSE: ORCL)
Oracle (NYSE:ORCL) carries the most direct OpenAI counterparty exposure. Q3 FY2026 RPO ballooned to $553 billion, up 325% YoY, anchored by a 10-gigawatt OpenAI partnership. The flip side: CapEx guidance of $50 billion in FY2026, trailing free cash flow of negative $24.7 billion, and non-current debt of $124.7 billion.
Reddit sentiment flipped from a bullish 62 on Sunday to a bearish 38 by Tuesday morning, citing the headline at issue: “OpenAI missed rev/user targets. $ORCL risk?” ORCL is down nearly 15% YTD already. Bull case: Management raised FY2027 revenue guidance to $90 billion, with prepayments and customer-supplied GPUs cushioning CapEx risk.
4. Micron Technology (NASDAQ: MU)
Micron (NASDAQ:MU) is the highest-beta name given memory cyclicality. Q1 FY2026 revenue was $13.64 billion, up 57% YoY, with the Cloud Memory Business Unit posting $5.28 billion at a 66% gross margin. The Q2 guide of $18.7 billion revenue and 68% non-GAAP gross margin reflects HBM3E sold out into 2027.
Reddit sentiment whipsawed from 85 (very bullish) on April 22 to 18 (very bearish) by April 24. Shares are still up 548% over one year. Bull case: HBM is the chokepoint of AI accelerator builds, with order books stretching into 2027. Bear channel: if hyperscaler GPU orders soften, memory bookings adjust faster than chips.
m
5. Corning (NYSE: GLW)
Corning (NYSE:GLW) reported Q1 2026 this morning: revenue of $4.14 billion, up 20% YoY, with Optical Communications at $1.85 billion, up 36%. UBS noted last December that the optical communications segment now accounts for over half of total core profit, driven by AI data center demand.
The stock has more than doubled over the past year, up 286%, after the Meta supply deal. Morningstar flagged it as overvalued. Bull case: management cited two new hyperscale long-term supply agreements similar in scale to the up-to-$6 billion Meta deal. Risk: Optical fiber demand is a derivative of hyperscaler buildout pace, not a leading indicator.
Conclusion
The common thread is hyperscaler CapEx concentration. Four of the five names trade at or near multi-year highs, with all citing hyperscaler AI spending as the primary growth engine. Reddit threads like “is anyone actually making money from AI or is it just the chip sellers?” now compete with bullish CapEx headlines. The revised Microsoft-OpenAI partnership announced April 27, which removes exclusivity and lets OpenAI sell across clouds, plus Alphabet’s planned $175–$185 billion in 2026 CapEx, complicate any clean bear thesis. Wednesday and Thursday earnings from Microsoft, Meta, Amazon, and Apple will show whether the CapEx curve is still bending up or flattening.
Top Gaining Stocks
Top Losing Stocks