Last week, the market decided Microsoft's AI spending was reckless, right up until Microsoft proved otherwise in a single session.

This week it's telling a similar story about Qualcomm, minus the happy ending so far.

Qualcomm fell about 7% on Wednesday after Apple's in-house modem chip showed up faster than anyone modeled. The stock hasn't stopped sliding since, and it's now sitting more than 40% below its 52-week high.

Not because the business fell apart. Because one relationship everybody already knew was ending is ending a little sooner than expected.

Here's what I want to work through today: whether that's actually a reason to sell Qualcomm, or a reason to look closer.

Missed Friday's Playbook on Microsoft's 15% rally?

Japan is expected to announce that Tokyo and Washington took joint action to support the yen after it fell to 40-year lows. Reuters reported this would be the first coordinated yen-buying effort since 2011, with Japan estimated to have sold nearly $59 billion to stabilize the currency. The bigger takeaway is that this is not just an FX story. A disorderly yen slide can feed inflation, rattle bond markets, and force policymakers into much more direct market intervention.

The Fool’s take is that Apple’s quarter was solid, but not enough to make the dip an obvious buy. iPhone revenue jumped 22% year over year to $54.3 billion, Mac revenue surged 29% to $10.4 billion, and total revenue rose 16% to $109.42 billion. The issue is valuation and headwinds, with memory costs, currency pressure, and softer services and China results keeping investors cautious despite the strong headline iPhone numbers.

Amazon’s latest results shifted the AI spending debate in its favor. AWS revenue grew 36.7% to $42.2 billion, its fastest growth in 18 quarters, while backlog reached about $496 billion and AI-related revenue crossed a $25 billion annualized run rate. The concern is still huge capex, with Amazon now expecting around $220 billion in 2026 spending, but this quarter finally gave investors clearer evidence that the AI buildout is translating into faster cloud growth.

If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.

This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.

Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.

Qualcomm dropped about 7% on Wednesday after fiscal third-quarter results, and the obvious read is simple. Apple is leaving.

Qualcomm's CEO said Apple-related chip revenue will fall roughly 50% between the September and December quarters, and by 2027 Qualcomm expects to supply zero iPhone modems. That's the headline everyone already knew was coming, just arriving faster than the models assumed.

Sell first, ask questions later is the easy trade here. It's also, I think, the wrong one.

The real question isn't whether Qualcomm is losing Apple. It already knew that was happening. The real question is whether what's replacing that revenue is actually good enough to matter.

Start with the numbers that didn't make the headline.

Automotive revenue hit a record $1.6 billion this quarter, up 61% year over year. Non-handset revenue, everything outside of phones, grew 28%, and management just raised its fiscal 2029 target for that business to $40 billion.

Qualcomm also signed a new deal to supply custom AI data center chips to ByteDance, its clearest step yet into hyperscale AI infrastructure. The licensing business, QTL, generated $1.28 billion in revenue this quarter at roughly 69% margin, a royalty stream that collects regardless of who makes the modem inside an iPhone.

That is not a company with nothing left after Apple. That is a company that spent years quietly building the exit ramp.

Now the real risk, not the overstated one.

The Apple decline is moving faster than Qualcomm itself expected, which means guidance could keep resetting lower for a few more quarters before it stabilizes. Q4 EPS guidance already came in at $2.05 to $2.25, below the $2.36 to $2.38 Wall Street wanted, largely on memory supply constraints tangled up with this exact transition.

Handset chip revenue overall fell 20% year over year to $5.1 billion, and that's still Qualcomm's largest single business by a wide margin. Automotive and data center are real, but neither is big enough yet to fully replace a shrinking smartphone franchise on its own.

Not a broken business. A business mid-transition, with a few uncomfortable quarters still ahead of it.

Qualcomm has gone from a 52-week high of nearly $260 down to around $148 today, a drawdown of more than 40% that has only gotten steeper since Wednesday's report. This wasn't a one-day accident. The stock was already down roughly 20% for the year heading into earnings, and the Apple guidance just gave sellers one more reason to keep pressing.

What's notable is how little support has shown up on the way down. Every level that held earlier this year, first near $180, then near $160, has given way without much of a fight, which tells you conviction among the remaining bulls is thin right now.

The next real test sits down near $130 to $135, a zone the stock hasn't traded at in years. If that breaks, this stops being a rough quarter and starts being a real repricing of what Qualcomm's smartphone-less future is worth. If it holds instead, that's a strong signal the selling is exhausting itself rather than accelerating.

Management guided to a roughly 50% sequential drop by the December quarter. Watch whether the actual number lands close to that or worse, since a miss here would confirm the transition is even messier than management admitted on the call. A number better than guided would be the clearest signal yet that this quarter marked peak fear.

A 61% growth rate on a $1.6 billion base is the kind of number that either becomes a real second pillar or turns out to be one unusually strong quarter. The next two prints will tell you which. This is the single number I'd weight heaviest over the next six months.

Qualcomm authorized a $20 billion buyback back in March. A stock this cheap, trading around 11 to 15 times forward earnings, is exactly the setup where a real repurchase pace should matter. If buybacks slow down here instead of accelerating into weakness, that's a signal management doesn't see the same value the numbers suggest.

Qualcomm's chip business is still tied to global smartphone demand, and China remains a meaningful swing factor outside of Apple entirely. Any further softness there would compound the Apple headwind rather than offset it, and it's the wildcard least under Qualcomm's own control.

I've watched a lot of "one customer is leaving" stories turn into permanent de-ratings, and I don't think this is one of them.

Qualcomm has known Apple was leaving for years. It spent that time building an automotive business growing north of 60%, a licensing arm that prints cash regardless of who wins the modem fight, and a real foothold in AI data center chips.

The stock is priced right now like none of that matters, at roughly 11 to 15 times forward earnings for a company still growing revenue and paying a rising dividend while you wait.

That gap between the price and the plan is exactly why I'm not selling. I'm watching for the moment the market catches up to what the numbers are already saying.

You don't have to scroll every AI thread, track every new tool, or watch every demo. 

The Rundown AI breaks it all down for you — the latest AI news, tools, and tutorials in one free 5-minute email every morning. 

Trusted by 2M+ professionals at Apple, Google, and NASA.

Sign up to start learning.

The scariest headlines are usually the ones investors already knew were coming.

Apple leaving Qualcomm wasn't a surprise. The timing was.

Markets often punish a known risk twice, once when it's announced as a future event, and again when it actually shows up, even if nothing about the underlying math has changed.

That gap between reaction and reality is where most of the good setups live.

The skill isn't predicting the headline. It's knowing which headlines were already priced in before you ever saw them.

If One Big Idea sharpened your week, send it to one friend who needed it too.