For most of this year, the market has treated memory chips as the scarcest asset in the entire AI trade.

Supply was tight. Pricing was locked in. SanDisk was the name that proved it, up 574% in 2026 alone.

Then on Monday, a report surfaced that the U.S. government might let Apple source memory from Chinese suppliers CXMT and YMTC.

No deal has been signed. No policy has changed.

The stock still lost $17 billion in market value in a single session.

One analyst is already calling it an overreaction. The tape, three days later, isn't so quick to agree.

Today I'm cross-examining both sides. SanDisk's bull case is loud, and it's backed by real numbers.

But the same scarcity that built this rally might be exactly what invited the policy risk that just hit it. That's the tension worth sitting with before you decide who's right.

Missed Monday's edition on CrowdStrike's earnings setup? 🎯 The One Number That Decides CrowdStrike's Wednesday

Boeing’s 737 MAX production ramp could become a second-order opportunity for aerospace suppliers. TransDigm and Woodward may benefit if higher Boeing output turns into stronger downstream demand for aircraft components, systems, and services. The caveat is execution: this upside only matters if Boeing can actually ramp production smoothly after years of supply-chain and quality issues.

Lego’s first-half revenue rose 21% to DKr41.9 billion, helped by strong demand across the Americas, Europe, and Asia Pacific, plus new product tie-ins including World Cup and Formula One ranges. The bigger takeaway is that Lego is still outgrowing the broader toy market by turning brand strength into pricing power, product relevance, and global demand across both children and adult collectors.

Micron’s AI-driven rally has been incredible, but investors should not forget how cyclical memory stocks can be. The bull case is still strong because data-center memory demand remains intense, AI infrastructure spending is huge, and Micron’s latest results showed massive revenue and cash-flow growth. But after a near-10x move, the margin for error is much thinner, so this looks more like a stock to buy carefully on pullbacks than chase blindly after the run.

AI agents are doing the work of full GTM teams. Pipeline, content, customer service. The startups using them aren't waiting for headcount approvals.

Get the free Practical Guide to Agentic GTM for Startups and see exactly where to start.

Unlock the Playbook

SanDisk shares fell 7.27% on Monday, closing at $1,480 and erasing roughly $17 billion in market value.

The market's confident case is simple. This was an overreaction to an unconfirmed report, and the AI memory shortage that built a 574% rally this year is still fully intact.

But a report doesn't need to become a signed deal to move a stock this violently.

So the real question isn't whether the report turns out to be true. It's whether SanDisk's story was ever as bulletproof as the last six months made it look, or whether the market just hadn't been tested yet.

Here's the case as the market has believed it since spring.

SanDisk went public at $52 in February 2025. By June 22 this year, it hit an intraday high of $2,354.39.

That's a gain of more than 4,400% in sixteen months. Almost nothing else in the S&P 500 has come close, and the move made SanDisk the index's single best performer this year.

Data center revenue surged 233% sequentially in a recent quarter as AI infrastructure buildouts demanded enterprise SSDs faster than the industry could make them.

Production cuts from the 2023 NAND downturn left supply structurally tight, and pricing power followed almost automatically.

Fiscal Q4 revenue came in at $8.965 billion, up 51% sequentially, with an 84.6% gross margin. That is not a normal quarter for a memory company. That is a business temporarily behaving like a monopoly.

Guidance for the next quarter calls for $10.3 to $10.8 billion in revenue, another step up.

Add index inclusions in the S&P 500, Nasdaq-100, and Russell Growth, and you get a stock that passive funds were forced to buy regardless of valuation.

Lynx Equity Research's KC Rajkumar called Monday's drop an "overreaction," pointing to "qualification gaps and constrained capacity" that keep Chinese suppliers from posing a near-term threat.

Wall Street's average target still sits at $2,126.17. That implies over 40% upside from Monday's close, and nobody big has walked that number back yet.

That's the bull case in full. Strong data, real margins, and analysts who aren't budging under pressure.

Now interrogate it.

The scarcity that built this rally is the same thing that just invited a government to consider loosening it.

Reports say the U.S. may allow Apple to source DRAM from China's CXMT and NAND from YMTC. Commerce Secretary Howard Lutnick opposed exactly this arrangement as recently as last week.

That a policy shift is even on the table is not a random headline. It's a signal that memory pricing has gotten uncomfortable enough for the biggest customer on earth to go looking for a workaround.

Not because SanDisk did anything wrong. Because it did something too right, for too long, at Apple's expense.

Here's the part that gets skipped in the "overreaction" framing. SanDisk was already down more than 30% from its June peak before this report even hit, as institutional money rotated out of memory names and into hyperscalers.

The stock's own momentum was cracking before the China headline gave it a reason.

That is the detail the bulls have to explain away, not just the rumor itself.

Susquehanna's price target is $3,250. Morgan Stanley's is $1,750. Same stock, same week, a $1,500 spread in what "fair value" even means.

This is not a market that has this priced with confidence. This is a market guessing, and guessing markets punish surprises hard, in either direction, without much warning.

There's a second layer here, and it has nothing to do with execution.

This isn't a competitive threat SanDisk can out-engineer its way past. It's a policy decision sitting on Howard Lutnick's desk, subject to politics SanDisk doesn't control.

If the Commerce Department greenlights the Apple-China sourcing arrangement, the scarcity premium in this stock doesn't erode slowly. It ends, potentially in a single announcement.

If it doesn't, SanDisk's multiple stays hostage to a headline risk that can resurface at any point between now and whenever Washington actually decides.

Either way, the company's own execution stopped being the only variable that matters here.

The chart tells a rougher story than the earnings deck does.

SanDisk went from a $52 debut to a $2,354 peak in under a year and a half, then spent the back half of summer giving a third of that gain back before Monday's headline even landed.

The June high now looks less like a launchpad and more like a ceiling nobody has tested since.

Monday's session opened calm, then broke down hard, touching an intraday low of $1,416.56 before buyers stepped back in and clawed the close up to $1,480.

That bounce matters. It means there is still real demand willing to defend this name in the $1,400s, not just sellers in control.

But the stock closed well below where it opened the week, at $1,596.08, and it's now trading closer to its post-peak lows than its post-peak highs.

If $1,416 gives way on the next leg down, there isn't an obvious floor until the market decides what this business is worth without a scarcity premium attached at all.

If it holds, the 574% rally still has a pulse. If it doesn't, the overreaction thesis needs a completely new argument.

Official word from Commerce. Nothing here becomes real until Howard Lutnick's office confirms or kills the Apple-China sourcing story outright. A firm denial in the next two weeks would validate the overreaction case fast, and probably fuel a sharp relief rally. Silence, or a soft confirmation buried in a policy briefing, would validate the opposite and keep the overhang alive well into next quarter.

NAND and DRAM spot pricing. If Chinese supply from CXMT or YMTC starts showing up in spot markets at real scale, it will hit pricing data weeks before it hits any earnings call. Watch spot price trackers over the next month, not just SanDisk's next print, since the market will react to the leading indicator first.

Micron's next earnings commentary. I haven't covered a memory name directly before this one, though I flagged the same AI infrastructure tension when AMD's data center revenue doubled and the stock still fell three weeks ago. Memory and AI compute are now trading on the same nerve. Micron's next update on capacity and China competition will move SanDisk whether or not SanDisk says a single word on its own call.

Whether $1,416 holds. This is the line in the sand from Monday's selloff, and it's the cleanest technical signal available right now. A decisive break below it on heavy volume would say the market believes the scarcity story is actually ending. A bounce and reclaim of $1,596 would say Monday was exactly the overreaction Rajkumar called it.

Analyst target revisions. Watch whether firms start drifting toward Morgan Stanley's more conservative $1,750 or toward Susquehanna's $3,250 over the next two weeks. A cluster of downward revisions would matter more to this stock than anything management says on the next call.

I don't think Monday was an overreaction, and I don't think it was the end of the story either.

Both can be true at the same time.

The bull case has real numbers behind it. 51% sequential revenue growth, 84.6% gross margins, guidance that keeps climbing quarter after quarter.

But a business doesn't get 4,400% from IPO to peak without pricing in a level of scarcity that was always going to attract a countermove, from customers, competitors, or governments.

That countermove just showed up three days before this newsletter went out, and the stock had already started cracking before it did.

I'd rather own this on the other side of $1,416 holding than in front of it.

The 574% headline is real. So is the fact that the easiest money in this trade was made before June, not after it.

Folks who are still buying the headline number instead of the chart are buying someone else's exit.

I'd want to see two or three sessions holding above $1,416 before I called this one resolved in the bulls' favor.

Creators lock in holiday content calendars 90 days out, before brands figure out commissions. Waiting too long to launch an affiliate program means less runway to build demand and a missed shot at the best partnerships.

The 90-Day Holiday Sprint covers commissions, recruiting, and scaling a program at Day 30, 60, and 90.

Download the Free Guide

The best businesses in history have all faced a version of this moment, the point where their own success becomes the thing that invites the fight.

Standard Oil got broken up because it was too good at what it did.

Big Tobacco got regulated because it was too profitable to leave alone.

SanDisk isn't in that league yet, and a sourcing rumor isn't an antitrust case.

But the mechanism rhymes across every scale.

Extreme pricing power rarely goes unanswered for long, whether the answer comes from a competitor, a regulator, or a customer that finally goes looking for a way out.

The lesson isn't to avoid scarcity trades altogether.

It's to remember that the scarcer something looks, the more attention it's quietly attracting from people who want that scarcity to end.

That's as true for a $17 billion memory stock as it is for anything else you'll ever own.

If this deep dive changed how you see SNDK, send it to one friend who's still trading the headline.