AI Demand Is Holding Up; The Market Is Testing the Payoff
A sharp tech-led selloff meets strong cloud and AI growth, while consumer and home-furnishings signals remain mixed.
The cleanest tell is the divergence between demand and the market’s price of risk
U.S. equities finished sharply lower Monday, November 16. At the 4:00 p.m. ET close, SPY was down 1.91% on the day and QQQ fell 2.66%; the quoted snapshots were from FMP, delayed 15 minutes. In the 30-day comparison ending November 16, QQQ was down 7.83% and technology-sector XLK down 11.40%, while consumer-discretionary XLY and retail XRT were each up just over 3%. That divergence says the current pressure is concentrated in growth and technology, not an indiscriminate rejection of every consumer-facing business.[1][2]
The day’s news points directly to the AI trade: Reuters reported that investors were bracing for Nvidia earnings and Anthropic’s market debut amid renewed semiconductor weakness, while AP described the broad decline as led by Nvidia. The headlines explain the immediate risk repricing; they do not establish that customers have cancelled AI projects or cut cloud consumption.[3]
That distinction matters for the hypothesis that earnings growth and resilient demand can support DDOG, SNOW, CRWV, AN, META, RH, WSM and ETH over the next year. The latest evidence supports a demand case in parts of software and AI infrastructure, but it does not validate the whole basket or remove capital-intensity, valuation and consumer risks.
Demand evidence is strongest in cloud software and AI compute
Datadog’s Q3 2026 revenue grew 35% year over year to $1.20 billion, according to its November 4 results release. On its earnings call, management said Q3 net revenue retention remained in the low 120s, gross retention in the mid-to-high 90s, and remaining performance obligations grew 38% year over year. This is evidence of usage and customer expansion, although RPO and retention are not guarantees of future growth.[4][5]
CoreWeave’s Q3 call described customers moving from experimentation to production; management cited more than 1.7 gigawatts of active power, expanding adjusted operating margins and over $250 million of booked ARR for its managed inference platform. Its Q3 results release characterized the quarter as record revenue with expanding margins. Those are tangible demand and execution signals, but the model also depends on rapid infrastructure deployment and financing. A contracted backlog is valuable only if capacity is delivered on time and customers pay as expected.[6][4]
Snowflake’s Q3 fiscal 2026 release reported $1.16 billion in product revenue, up 29%, 125% net revenue retention and $7.88 billion in remaining performance obligations, up 37%. Those figures are supportive, but the source labels that release as fiscal Q3 2026 even though it is dated from the prior fiscal-year reporting cycle. Because of that period-label ambiguity, treat it as a company-reported operating datapoint, not a clean read-through to the latest quarter in November 2026.[7]
Meta shows the return-on-investment question; consumers are not one trade
Meta’s Q3 results beat revenue estimates, but the market’s response focused on the cost of the AI build-out: CNBC reported that free cash flow turned negative and the company raised the lower bound of its 2026 capital-expenditure outlook. Strong advertising revenue can finance AI investment, but the next question is whether that spending produces durable incremental ad performance, products or revenue at acceptable returns.[8]
For AutoNation, Q3 adjusted EPS rose to $5.11 from $5.01 a year earlier, and the company highlighted record after-sales gross profit and 6% customer-pay growth. Yet third-quarter revenue was roughly flat, while new-vehicle sales and reported profit declined year over year. That is resilience in service and finance, not proof that all parts of vehicle demand are strengthening.[9]
Home-related evidence is more uneven. RH’s latest reported quarter before this snapshot was Q2: revenue grew 2.6% year over year, but its next-quarter revenue outlook was reported below analyst expectations. Williams-Sonoma’s Q2 comparable-brand revenue rose 6.2% and the company raised its full-year outlook, while the market was awaiting its next report. Ethan Allen (ETD, not ETH) reported fiscal Q1 2027 sales down 11.8% year over year and net income down 83.9%. These contrasts argue against treating home furnishings as a single broad demand signal.[10][11]
| Thesis component | Evidence in favor | What keeps the case conditional |
|---|---|---|
| DDOG / SNOW software demand | Growth, retention and contracted-work indicators remain substantial | Cloud consumption can fluctuate; reported periods and forward conversion matter |
| CRWV AI infrastructure | Fast growth, production adoption and a growing inference business | Heavy build-out creates delivery, financing and return-on-capital risk |
| META monetization | Strong Q3 revenue and a large advertising business | AI capex and negative free cash flow raise the hurdle for returns |
| AN / RH / WSM / ETD consumer demand | Auto after-sales and WSM Q2 were relatively resilient | New-vehicle revenue/profit, RH outlook and ETD sales all show softer pockets |
| ETH / Ethereum | A distinct risk asset that may respond to liquidity and risk appetite | It is not an operating company; it should not be grouped with earnings growth |
Macro context is supportive on growth, less so on discount rates and sentiment
The latest macro snapshot available here is October 2026: unemployment was 4.1%, real GDP growth was 2.07% year over year and industrial production growth was 2.25%. But the 10-year Treasury yield was 5.1%, consumer sentiment was 48.1, and the VIX stood at 18.7, up about 23% month over month. That combination is not a recession signal in this dataset; it is a mixed backdrop where ongoing activity coexists with expensive long-duration discounting and weak household sentiment.[12]
Ethereum (ETH) is not a public-company earnings ticker. A current web snapshot put ETH near $1,819 on November 16, but a separate quote feed returned an implausible $17.36 for the symbol, indicating a ticker/data mapping problem. We therefore do not use that quote feed for ETH or treat a point estimate as verified here. Its role in the thesis should be evaluated separately from company-level earnings and operating demand.[13][1]
What would change the balance of evidence?
The constructive reading would gain support if cloud consumption and retention stay firm, CoreWeave converts contracted capacity into cash-generating operations without excessive financing strain, and Meta demonstrates returns on the higher AI investment. A more cautious reading would gain support if new spending or backlog fails to convert into recognized revenue, profitability weakens as infrastructure scales, or consumer-sensitive names report further demand deterioration.
What to watch next
- In cloud software, look for customer expansion, usage trends, and whether contractual obligations turn into revenue.
- In AI infrastructure, compare new capacity delivery and utilization with capital needs, operating margins and cash generation.
- At Meta, focus on advertising monetization alongside the size and expected payoff of AI spending.
- For AN, RH and WSM, separate service/after-sales strength from vehicle and discretionary home-purchase demand; compare each company’s next reported results with its own guidance.
- For Ethereum, use crypto-market and liquidity evidence, not corporate earnings metrics; verify the data source before citing a current price.
The current evidence does not make the eight-name thesis one-sided. It supports real demand in selected software and AI-infrastructure businesses, but Monday’s selloff is a reminder that strong demand alone may not settle questions about execution, financing, valuation or the consumer cycle. This is research, not a forecast or a recommendation.
Sources
- Quote: SPY
- Compare stock charts
- Morning Bid: AI selloff resumes as Nvidia earnings loom | Reuters
- AI Demand vs. Consumer Demand: What Q3 Shows | FN2 Research
- Datadog, Inc. (DDOG) Q3 FY2025 2025-11-06T08:00:00
- CoreWeave, Inc. Class A Common Stock (CRWV) Q3 FY2025 2025-11-10T17:00:00
- Snowflake Reports Financial Results for the Third Quarter of Fiscal 2026
- Meta - Meta Reports Third Quarter 2026 Results
- AutoNation, Inc. - AutoNation Reports Third Quarter 2026 Results
- Williams-Sonoma faces earnings test as freight costs pressure margins By Investing.com
- Ethan Allen Reports First Quarter Results :: Ethan Allen Interiors Inc. (ETD)
- FRED: Unemployment
- ETH | Currently at $1,819.01 (+2.84%) | Nov 16, 2026