Jim Cramer's top 10 things to watch in the stock market Thursday

The article is broadly constructive on AI and tech spending, highlighting Cerebras' $5.5B IPO, Cisco's blowout quarter and ~15% premarket gain, and multiple bullish price-target hikes on Dell, Broadcom and Apple. It also notes Starbucks upgrade support, while Doximity is the major negative with shares down over 23% after a weak fiscal 2027 forecast and a wave of downgrades. Geopolitically, the Trump-Xi summit headlines are described as mostly positive, but Taiwan remains a sticking point.

Analysis

The strongest read-through is that AI capex is shifting from a narrative trade to a budget-line reality: networking, power, and inference-oriented compute are getting funded even as scrutiny rises on model economics. That favors the picks-and-shovels layer with the least customer concentration risk — network gear, server integrators, and specialized chip designers — while increasing pressure on businesses that depend on broad enterprise IT demand to justify premium multiples. The secondary winner is upstream memory and high-speed interconnect suppliers; if server refresh cycles accelerate, component inflation becomes the next margin battleground, not demand.

Cisco’s upside matters more as a signal than as a stock-specific event. If hyperscalers are still ordering aggressively, the risk is not a demand cliff but a supply-chain constraint that shifts bargaining power toward vendors with execution credibility and away from laggards. That also supports Broadcom’s networking franchise and keeps Nvidia’s inference stack protected, but it raises the probability that enterprise buyers will experience longer lead times and higher total system costs over the next 2-3 quarters.

The biggest downside setup is in names with weak forward guidance and fragile estimate support. Doximity’s reset is a reminder that adjacency to a strong vertical does not immunize a company from multiple compression once growth decelerates; in this tape, “AI exposure” is not enough unless it translates into monetization and durable operating leverage. On the other side, Starbucks and Apple look like slower-burn compounders where the market may still be underestimating the duration of operating improvement and services/margin mix, especially if rates stabilize.

Contrarian view: the market may be over-rotating into AI hardware winners while underpricing the eventual margin squeeze from memory, networking, and power bottlenecks. The trade is not to fade the theme, but to own the strongest balance sheets and pricing power within it, because the second-order effect is that capex growth can stay high even while incremental returns on that capex fall. The next catalyst to watch is whether enterprise demand broadens enough to offset hyperscaler concentration; if not, the winners will be narrower than current consensus assumes.

Key Decisions for Investors

• Long AVGO vs short DOCS for 1-3 months: favor a beneficiary with durable AI/networking exposure over a guidance-reset name where multiple compression can persist after estimate cuts.
• Add to CSCO on pullbacks over the next 1-2 weeks: use the post-earnings demand validation as evidence that networking remains a bottlenecked spend category; downside is limited unless hyperscaler capex pauses.
• Initiate a pair trade long DELL / short a weaker server OEM basket for the next earnings cycle: Dell has pricing agility and enterprise breadth, while component inflation should punish less disciplined players first.
• Buy AAPL on any market-wide AI rotation fade over the next 2-4 weeks: this is a lower-beta compounder with optionality from mix and services; risk/reward improves if rates remain range-bound.
• Avoid chasing SOLS after the separation rally; if keeping exposure, size it as a trailing stop position only, because post-spin multiples often mean-revert once the initial scarcity premium fades.