Nvidia Earnings & Fed Minutes Test Bearish Sentiment

Good morning investors,

ChatGPT went down early yesterday and for a moment you would have thought the world had ended. The silver lining was a blessed reduction in posts stuffed with dramatic em dashes, so perhaps outages have their place in the universe after all.

Today the spotlight is on Nvidia, the Fed, and the US consumer. Markets have sold off for four straight sessions into this earnings print, bitcoin has retested the 90,000 level, and AI is now being treated as both the biggest opportunity and the biggest tail risk in the system.

Yesterday’s session extended the losing streak to a fourth day, with the S&P 500 ($SPY ( ▲ 0.23% )) down about 0.8%, the Dow ($DIA ( ▲ 1.36% )) off 1.1%, and the Nasdaq ($QQQ ( ▼ 0.32% )) lower by 1.2% as investors continued to de-risk ahead of Nvidia’s ($NVDA ( ▼ 1.55% )) earnings and the delayed September jobs report. Tech once again led the decline as crowded AI names sold off.

This morning, futures are trying to stabilize. Dow futures are modestly higher, S&P 500 futures are up around 0.3%, and Nasdaq futures are up about 0.4% as traders position around tonight’s results. The tone is more “braced for impact” than bullish, but at least the market has stepped away from the panic button.

Bitcoin briefly broke below 90,000 dollars yesterday before bouncing, and is currently trading around 91,000 to 92,000 dollars, still well off the October peak but not in free fall. Gold has picked back up as investors hedge both equity and policy risk. In other words, positioning is nervous, not complacent.

Nvidia earnings tonight remain important, while still directionally determinative for the entire market, I do think the actual event is less important than in the past, as we have seen a lot of their data from other reports. Cloud and hyperscaler capex plans tell you where AI demand is going months before the semiconductor numbers show up in a quarterly print. Nvidia has become more of a high beta confirmation tool than the steering wheel.

I still expect a solid beat and a constructive outlook. Most of what matters for this quarter is already known through the lens of cloud budgets, prior mega-cap earnings, and the ongoing AI infrastructure build out. A positive surprise is possible, but my base case is a high-quality beat with minimal genuine “new” information.

If the stock can reclaim and hold key resistance in the 210–215 dollar area over the coming sessions, that would help repair sentiment even if the narrative does not change fundamentally. Right now Nvidia is functioning more as a sentiment barometer than a fundamental driver. The options market is still pricing an enormous potential move, but that says more about recent volatility and positioning than about the company suddenly becoming opaque.

Jensen Huang’s pre-scheduled equity sales sparked predictable outrage in recent weeks despite being formulaic and pre-planned. Markets chronically misread insider transactions during fragile sentiment periods, turning small portfolio management decisions into big stories. The underlying demand picture does not rise or fall on whether Huang files a Form 4 in a choppy week.

Where this print really matters is at the margin of the AI bubble debate. Bank of America’s latest fund manager survey has a potential AI bubble as the number one tail risk, with managers also saying for the first time in two decades that companies are over-investing in capex. Low cash balances in that survey are a classic warning sign. Even a strong Nvidia quarter could be used by skeptics as “proof” that AI capex has gone too far too fast, at least tactically.

I still see tonight as more of a sentiment catalyst than a regime shift. A strong beat and confident guide can trigger a relief rally and support a year-end bounce. A miss or cautious tone can extend the current flush. In both scenarios the structural AI case remains intact, but the price you are willing to pay for it changes.

While investors argue over whether AI is in a bubble, the underlying technology and capital cycle are still accelerating.

Google’s launch of Gemini 3 is a genuine step change. This is not a mild fine-tune, it is a rebuilt Mixture of Experts transformer with native multimodality, million token inputs and sparse activation that lets Google run frontier-level reasoning without instantly turning enterprise cloud bills into horror stories. Benchmarks across reasoning, code, multi-step planning and interface understanding show meaningful jumps, pushing models closer to truly agentic behavior rather than simple call-and-response chat.

At the same time, Anthropic’s latest funding round, with roughly 15 billion dollars of fresh commitments and a valuation stepping up toward 350 billion, locks in tens of billions of dollars of future compute spend with Microsoft and Nvidia. Microsoft is plainly diversifying its model bets beyond OpenAI, and Nvidia is embedding itself deeper into the model stack by financing the very customers that will keep buying its hardware.

Those two developments matter more to the long-term AI story than one quarter of Nvidia earnings. They tell you that the AI build out is still in the early innings in terms of infrastructure, that the capex cycle is enormous, and that a handful of platforms will likely dominate both model performance and economics. That is why AI can feel like a bubble and a necessity at the same time.

Away from AI, the consumer is quietly telling you to dial down the growth assumptions.

Target posted another soft quarter. Adjusted EPS of 1.78 dollars beat expectations, but revenue of 25.27 billion dollars missed slightly and comparable sales fell 2.7 percent year on year. Traffic declined 2.2 percent, average ticket slipped 0.5 percent, and management trimmed the top end of full-year EPS guidance to a 7.00–8.00 dollar range. Holiday sales are expected to decline low single digits year on year.

The mix tells the story. Essentials, beauty and food are holding up. Discretionary categories are weak. Even in seasonal moments such as Halloween, customers bought candy in volume but skipped décor. The pause in SNAP benefits during the shutdown added pressure at the margin, but the bigger issue is a cautious consumer who is unwilling to spend on “nice to have” items when uncertainty remains high.

Incoming CEO Michael Fiddelke is already in rebuild mode. Eliminating 1,800 corporate roles, lifting capex by 25 percent to around 5 billion dollars, and pushing for a simpler assortment, sharper design and better store execution are the right strategic moves. The problem is that Target is trying to fix complexity, execution and brand positioning at the same time, after four years of flat sales. Execution risk is real.

Lowe’s sits on the other side of the consumer but echoes the same theme. The company delivered a slight earnings beat at 3.06 dollars per share, revenue of 20.81 billion dollars and a 0.4 percent rise in comparable sales, but missed Wall Street’s comp expectations and saw net income decline to 1.62 billion dollars from 1.70 billion a year ago. Full-year sales guidance has been lifted to 86 billion dollars thanks to the Foundation Building Materials acquisition, but EPS guidance has been tightened to about 12.25 dollars as management acknowledges the macro headwinds.

The housing backdrop is still tough. Slow turnover, elevated mortgage rates and deferred big-ticket projects continue to weigh on both DIY and pro spending. Lowe’s strategy is to lean harder into the professional segment and squeeze synergies from acquisitions. That is a rational response, but it relies on sustained execution rather than an easy macro tailwind.

Walmart reports tomorrow and will be the key cross-check. If Walmart confirms that lower and middle income consumers are still trading down and prioritizing essentials, it reinforces the message from Target and Lowe’s that the US consumer is not broken, but is highly selective and very price sensitive heading into the holiday period.

Monetary policy is back in focus from two angles.

First, President Trump has made clear he already knows who he wants to nominate to replace Jerome Powell as Fed chair, hinting at a shortlist that includes both “standard” names and more surprising options. The list reported so far includes current governor Christopher Waller, governor Michelle Bowman, former governor Kevin Warsh, White House NEC director Kevin Hassett and BlackRock’s Rick Rieder. The prospect of a new chair with a different tolerance for inflation and growth risk is another layer of uncertainty for markets that are heavily invested and already nervous about valuations.

Second, the Fed’s October minutes are due today. They are likely to reveal a more divided committee than the simple “cuts are coming” narrative implies. Officials are dealing with an official data blackout, a delayed jobs report and tighter funding conditions in the repo market even after recent easing. High overnight funding costs heading into year end suggest that liquidity is not nearly as abundant as the policy rate alone would imply.

Overlay this with Bank of America’s survey showing low cash levels and AI bubble concerns, and you have a market that is long risk, short cash and reliant on the Fed to validate current positioning. That is not a comfortable place to be. Markets are less afraid of bad news than of a central bank that looks uncertain.

Bitcoin’s move below 90,000 dollars yesterday, followed by a rebound back above 91,000 dollars, is consistent with past periods where the market de-risks ahead of a big catalyst and then tries to stabilize. It is still a high beta expression of risk appetite rather than a safe haven. For me the key questions are whether institutional flows into products like spot ETFs remain constructive and whether we see signs of forced selling, which so far has been limited.

More broadly, we are seeing a rotation out of the most crowded AI and tech names and into relatively defensive or neglected sectors, with health care holding up better. That fits the survey data on perceived bubbles and over-investment. It argues for selectivity rather than abandoning growth altogether.

On the economic side, we have MBA mortgage applications, October housing starts and building permits, alongside international trade data. These will help flesh out the picture painted by Home Depot and Lowe’s about housing and home improvement demand.

On the corporate side, Nvidia, Palo Alto Networks, TJX, Williams-Sonoma, Bullish and Jack in the Box all report today, with Walmart tomorrow completing the retail picture. Tesla’s Arizona ride-hailing permit, Boeing’s Dubai Airshow order recovery, Brookfield’s 10 billion dollar AI infrastructure fund, Meta’s privacy investigation in Spain and Constellation’s Three Mile Island restart funding all add texture to the theme that AI, power and regulation are now tightly linked.

Four consecutive down days into a headline earnings event always feel worse than they are. Nvidia’s report tonight has been framed as a make-or-break moment for AI and, by extension, for the broader market. I do not see it that way.

Most of what matters for Nvidia is already visible in cloud budgets, AI product launches and the capital commitments from companies like Google, Microsoft and Anthropic. I expect a solid beat and constructive guidance, with limited genuine surprise. That can still be a positive catalyst if positioning is cleaner and expectations have cooled, but it is unlikely to rewrite the AI story overnight.

If Nvidia rallies on good numbers, it supports the case for a year-end bounce and validates using this pullback to add selectively to quality names. If it sells off on “good but not good enough,” it probably accelerates the flush and gives patient investors better entry points. In both scenarios, the work for individuals is the same: stay focused on balance sheets, cash flow and durable competitive advantages rather than on a single earnings call.

Use the volatility to upgrade what you own, avoid leverage, and resist the urge to rebuild your entire plan because someone on television says “AI bubble” with enough conviction. Their book and yours are rarely the same.

Please feel free to reach out to me on LinkedIn or by email if you would like help navigating this market environment or have any planning related questions.

Dan Sheehan

[email protected]

This newsletter is for informational purposes only and should not be considered as investment advice. Please consult with your financial advisor about your specific situation.