AVGO, NVDA, MU: Chips Trade Like the Cycle Is Eternal

Published on: Oct 28, 2026
Author: Brandon Kwan

Broadcom just got yanked from the “dead money” bin and thrown back onto the hot seat. The catalyst is not some mystical AI halo or a caffeine-fueled buyback rumor; it’s a fresh analyst upgrade tied to giant customer commitments that have the market suddenly re-pricing what “quality” means in semis. And because this is Wall Street, the same people who spent the year pretending the AI trade was mature are now looking at it like it has another lap left.

The setup is simple: when one chip name moves, everyone starts acting like they discovered the future in a spreadsheet. Broadcom is the headline act, but the real action is in the broader semiconductor pack, where investors are deciding whether this is a victory lap or just the pregame for the next leg higher.

Broadcom leads the chip rerating

1. Broadcom (AVGO)

Broadcom was the name doing the heavy lifting on Tuesday, rising 4% after D.A. Davidson analyst Gil Luria upgraded it to Buy from Neutral and lifted his price target to $400 from $350. The upgrade leaned on Anthropic’s S-1 filing and the expectation that OpenAI will follow with its own IPO next year. Luria’s read is blunt: the market is pricing Broadcom as if those IPO-linked opportunities barely exist.

Trading profile: Broadcom closed Monday at $329.17, so the $400 target implies 21.5% upside. Even after Tuesday’s move, the stock is still down 2% year to date, while the PHLX Semiconductor Index is up 60%. That gap is the kind of spread that makes portfolio managers feel either brilliant or late, sometimes both before lunch.

Key takeaway: The market is no longer just valuing Broadcom as a chip supplier; it is starting to price it as a structural AI infrastructure lever with customer visibility that stretches years out. If the market takes the S-1 disclosures seriously, Broadcom’s so-called “gone nowhere” year starts looking more like a setup than a verdict.

2. Nvidia (NVDA)

Nvidia is in the same conversation because Luria used it as part of the “biggest dislocation in the market” he sees between major chip makers and the rest of the AI hardware trade. His point is that companies like Nvidia, Broadcom, and Micron are trading as if the cycle is already over, while other AI-exposed hardware names are still priced as if the fun could last forever. That is not exactly a gentle take; it is a public warning that the market is misreading how durable the AI buildout may be.

Trading profile: No fresh price move or volume figure was provided here, but Nvidia remains the kind of bellwether that drags sentiment across semis whenever the tape wants a story. In practical terms, it is still the stock investors use as a lie detector for AI demand. If Nvidia stays bid, the whole sector gets to keep pretending this is still a growth trade and not a very expensive utility with better branding.

Key takeaway: Luria’s view suggests Nvidia is not being singled out as weak but as one of the names the market is underestimating if the AI spending cycle extends. For investors, the message is less about chasing headlines and more about recognizing that the market may still be underpricing the staying power of the biggest infrastructure winners.

3. Micron (MU)

Micron also showed up in Luria’s list of stocks the market is treating as if the cycle is already over. That matters because Micron is not usually the stock people buy when they want poetry; it is the stock they buy when memory pricing and capex discipline start to look less like a spreadsheet chore and more like leverage to the next phase of the AI buildout. Luria’s line was that there is a dislocation between names like Micron and the rest of the AI hardware crowd.

Trading profile: The evidence pack gives no fresh move for Micron, but its inclusion in the analyst’s framing tells you where it sits in the trade: not as a flashy AI darling, but as a cyclical chip name whose valuation can expand quickly when investors stop assuming the cycle is done. That is the kind of stock that can go from ignored to indispensable without changing much except the narrative.

Key takeaway: Micron remains the “show me” name in the group, but Luria’s argument is that the market is already too pessimistic on the duration of the cycle. If investors start believing the AI hardware spend lasts longer than expected, Micron should not be left off the list of names that can benefit from a broader rerating.

4. Anthropic

Anthropic is not a public stock, but it is now part of the Broadcom trade whether people like it or not. Its S-1 filing disclosed five-year master equipment leases with Broadcom worth about $161.2 billion, excluding interest. That number is doing a lot of work here. It gives the market a concrete reason to believe Broadcom’s AI revenue machine may be larger and stickier than skeptics assumed.

Trading profile: There is no listed share price to punch into a model, which is always convenient when a private company becomes the emotional center of a public equity trade. Still, the filing itself is what turned the volume up on Broadcom, because it put a real dollar figure on future infrastructure demand. The market loves a big round number almost as much as it loves pretending it discovered it first.

Key takeaway: Anthropic’s filing gives Broadcom something the market can actually underwrite: long-duration customer demand tied to equipment deployment. For investors in the public semis, that means the AI story is not floating on vibes alone; it now has lease commitments with enough scale to make the cynics at least do the math.

5. OpenAI

OpenAI is the sequel the market is already gaming out. The current thesis is that OpenAI is expected to follow Anthropic with its own IPO next year, and Luria argues that such a development would help Broadcom meet its multi-year targets. In other words, Broadcom’s upside case is no longer just about one customer or one filing; it is about a pipeline of public-market validation for the AI buildout.

Trading profile: Like Anthropic, OpenAI is not a listed stock in this context, but it matters because the market is trying to translate private AI demand into public equity multiples. That is the kind of thing that makes sell-side analysts sound either visionary or dangerously caffeinated. The actual point is straightforward: if more AI customers turn public, more of their spending becomes easier for investors to model, admire, and bid up.

Key takeaway: OpenAI is the future-proofing argument for Broadcom bulls. If it goes public next year, it could reinforce the view that the AI capex cycle still has room, which is exactly the sort of sentence that sends semis traders back into their natural habitat: overconfident, under-caffeinated, and suddenly very interested in forward multiples.

Investor Lens

The clean read is that Broadcom is no longer trading like a sleepy infrastructure name; it is trading like a beneficiary of a much longer AI spending runway than the market wanted to admit. Luria’s upgrade, the Anthropic disclosures, and Broadcom’s own revenue outlook all point in the same direction: the street may have been too eager to declare the cycle done.

The broader lesson for investors is less romantic and more annoying. When the market decides the AI capex story still has legs, the winners are not just the headline names; they are the suppliers with enough leverage to turn customer ambition into actual revenue. That is where Broadcom, Nvidia, and Micron stop being separate tickers and start looking like one very expensive bet on the same future.

AI