NVIDIA Rises 4% to $175.23 Despite Baird Cut and Huang Cloud

NVIDIA stock climbed 4% to $175.23 on November 11, staging its biggest one-day gain since late August even as Baird cut the chipmaker to Neutral from Outperform. The rally follows a two-day selloff that had pushed the stock near its 52-week low. The move matters because it tests whether Wall Street's fading conviction on AI hardware can outweigh evidence that demand still exceeds supply.

The context is a downgrade with unusual teeth. Baird didn't just lower its rating — it slashed its price target to $200 from $500 and said it will hold off on publishing 2027 data center forecasts for NVIDIA entirely. Morgan Stanley and Bank of America have also cooled on the stock. Buyers stepped in anyway after a beat-and-raise quarter from CoreWeave (NASDAQ:CRWV), the AI cloud infrastructure provider, revived the AI trade before the open.

The rebound is broad. Advanced Micro Devices (NASDAQ:AMD) rose 5% to $541.28, beating NVIDIA's gain. Broadcom (NASDAQ:AVGO) climbed 4% to $345.83, a slightly firmer advance than NVIDIA's. Both are rising without a fresh analyst downgrade hanging over them.

Why did Baird downgrade NVIDIA despite strong demand?

Baird's reasoning rests on visibility, not demand destruction. The firm said visibility has deteriorated enough that it will not publish 2027 data center forecasts for the company. It sees supply constraints as the main risk to NVIDIA's growth, with softer demand as a possible contributor. Slower growth in AI spending by large cloud companies in 2027 also figures into Baird's outlook.

Leadership weighed on the call too. Baird cited uncertainty over whether CEO Jensen Huang stays on into 2027, after reports about his personal conduct and a board investigation. Prediction markets now put the odds of Huang leaving at 15%, per a separate report. Even so, Baird acknowledged the core fact bulls cling to: demand for NVIDIA chips still exceeds supply.

Morgan Stanley made its own adjustment under a stop-loss policy for top picks. Analyst Joseph Moore removed NVIDIA as the firm's top semiconductor pick after the stock dropped more than 25% since October. Morgan Stanley kept its Overweight rating and $300 price target. Bank of America removed NVIDIA from its U.S. 1 List on November 10.

The takeaway: three major banks have trimmed their stance in weeks, yet none of them — including the downgrade's author — disputes that NVIDIA can't make chips fast enough.

How did CoreWeave reignite the AI trade?

CoreWeave's beat-and-raise quarter lifted chip, memory and data center names, along with NASDAQ 100 futures, before the open on November 11. Goldman Sachs analyst James Schneider called AI stocks "clearly in oversold territory" and named NVIDIA, Broadcom and AMD as top picks.

Schneider laid out a valuation case. He said NVIDIA trades at 12x and Broadcom at 13x a no-growth 2028 baseline — a scenario he considers unlikely. He called AMD "very attractive" for growth, though less compelling on valuation. The caveat: Schneider warned that catalysts for a sustained rally in chip stocks like NVIDIA may be limited before year-end.

That is the tension in this rally. The multiples look cheap against a worst-case 2028, but nobody — not Schneider, not Baird — has identified a near-term catalyst that would confirm the bullish case before the calendar turns.

How do NVIDIA, AMD and Broadcom sell into the AI buildout?

The three companies monetize the AI buildout in different ways, and the differences explain why they trade together but not identically.

• NVIDIA sells a full platform: GPUs, networking, systems and the CUDA software ecosystem that ties them together. On the company's August 26 earnings call, Huang said: "NVIDIA is a platform, an entire AI factory platform that spans the entire AI life cycle, that you can use in any cloud."
• AMD pairs its Instinct GPUs with EPYC server CPUs in Helios rack-scale systems and backs them with its open ROCm software. It also runs PC, gaming and embedded businesses, giving it a broader revenue mix than NVIDIA.
• Broadcom co-designs custom accelerators with AI labs such as Anthropic and OpenAI and sells the Ethernet networking that links large AI clusters. VMware-based infrastructure software adds an enterprise revenue stream the GPU makers lack. CEO Hock Tan argued on the company's September 2 call that a chip co-developed for specific large language model workloads "will outperform any GPU."

The takeaway: NVIDIA sells the standard platform, Broadcom sells custom silicon plus networking, and AMD sells a diversified alternative — three bets on the same buildout with different risk profiles.

What does NVIDIA's November 17 report have to prove?

NVIDIA reports Q3 FY2027 results after the close on November 17. Shareholders want reassurance that the AI investment cycle is durable. The specific test is whether results back up the $108 billion revenue outlook the company issued in August in its SEC filing.

The macro backdrop complicates the picture. Chicago Fed President Austan Goolsbee called the latest inflation report "borderline disturbing" while the Federal Reserve weighs more rate hikes. Tighter policy historically pressures high-multiple growth stocks hardest.

The bull case has real substance. Baird's own finding that demand exceeds supply supports it, as does Schneider's low-multiple argument. The same expansion thesis extends to the power, cooling and networking suppliers that feed the data center buildout beyond the chipmakers themselves.

The bear case is equally concrete: an unresolved leadership question at the top of the company, slower expected cloud spending growth in 2027, and a Fed leaning toward tighter policy. A cautious approach may be warranted until the Huang question settles — a board investigation with a 15% implied exit probability is not a detail the market can price cleanly.

November 17 is the deadline for evidence. Until then, the rally rests on discounted multiples and one strong CoreWeave quarter, not on answers to the two questions — Huang's tenure and 2027 cloud budgets — that triggered the selloff in the first place.