Here are the top 10 things we're watching in the stock market Friday

The session is framed by a rate-driven pullback with bond yields steady, but stock-specific catalysts dominate: Broadcom is reportedly in talks to help raise $60B in debt for an AI chip financing deal (via Bloomberg), and multiple AI/semi names are being initiated/covered with buy ratings (e.g., Micron initiated at $1,300 on a “memory supercycle”). Retail earnings/guidance are mixed—Walmart shares are flat after a 9% drop and price-target cuts following disappointing results, while Ross Stores is up >8% on a beat and 10% comparable sales, and BJ’s Wholesale is up ~2.5% with comps ex-gas +3.1% and a raised full-year outlook. Abbott also agreed to pay $670M to resolve preterm infant formula litigation, removing a long-standing overhang.

Analysis

The AI-financing angle matters less as a one-off headline than as a signal that the ecosystem is trying to de-risk demand destruction from higher capex. That is constructive for AVGO and NVDA near term because it widens customer purchasing power, but it also introduces a hidden credit filter: if customer-level economics tighten, the weakest projects get financed last, not first. Over 1-3 months the market will care whether these structures actually close and whether private-market valuations for AI apps, including the upcoming Anthropic IPO, validate the spend; if those do not clear, the bullish read-through to semis and banks fades fast.

The more tradable dispersion is in retail. ROST’s strength says off-price is still taking share, while TJX looks like a later-cycle catch-up story rather than an immediate winner; that gap can persist until inventory metrics converge. WMT’s reset is different: it is not a traffic problem so much as a margin re-rating problem, and that makes it vulnerable if pricing pressure stays intense or if the consumer weakens enough that it cannot offset mix deterioration with volume.

On the industrial side, ETN and GEV remain cleaner expressions of the AI buildout than the semis because power bottlenecks are the binding constraint, not chip demand. The setup is vulnerable to higher yields: if rates back up, long-duration AI and data-center multiples can compress even while backlog stays intact. ABT’s settlement mainly removes a litigation overhang, but unless reserves or future claims step down materially, the earnings impact is more optical than fundamental.

Key Decisions for Investors

• Long ETN and GEV on any rate-driven pullback over the next 1-3 months; this is the cleaner AI-capex expression with less dependency on customer financing execution than AVGO/NVDA. Falsifier: a sustained move higher in 10Y yields or evidence that hyperscaler capex is being delayed, not accelerated.
• Pair trade: long ROST / short TJX for the next 4-8 weeks. ROST has clearer estimate-upgrade momentum from comp outperformance; TJX still looks like a catch-up story, which is usually lower quality after the first leg. Exit if TJX shows two consecutive weeks of improving weekly checks or if the spread narrows materially ahead of earnings.
• Tactical short WMT into strength only if the stock retraces part of the post-earnings gap and the market starts rewarding margin repair over traffic defensiveness. Risk/reward is asymmetric only if price investment continues to pressure operating income; stop if management signals a faster-than-expected recovery in merchandise margin.
• Buy AVGO only on confirmation that the financing framework converts into actual committed demand, not just strategic publicity. Use it as a 1-3 month momentum trade, not a long-duration thesis; the key risk is that debt support substitutes for organic cash generation and the market eventually discounts the structure.
• No urgent trade on ABT from the settlement alone; treat it as an overhang removal, not a re-rating event. Reassess only if future quarterly reserve language steps down or if the litigation discount closes materially versus peers.