#184 Intel's Fall from Grace

Fresh & Hot curated AI happenings in one snack. Never miss a byte 🍔

This snack byte will take approx 4 minutes to consume.

Since its inception in 1968, Intel has been synonymous with shrinkage. Back then, though, it was something to brag about. Every couple of years, the chip giant would unveil transistors half the size of the previous generation, ushering in what we now call Moore’s Law, named after one of Intel’s founders.

This magical shrinking allowed twice as many chips to fit on a single silicon wafer, meaning more chips sold for roughly the same price—pure genius.

Intel quickly cornered the market for memory chips and later dominated the microprocessors that fueled the PC revolution. Ah, those were the golden days.

Fast forward to today, and when you hear "Intel" and "shrinking" in the same sentence, it's not quite the compliment it used to be. After a brutal couple of quarters, Intel's market value has shrunk to a paltry $84 billion, down from over $210 billion in January. Ouch.

While the AI boom is making other chip companies soar, Intel’s stock is as cheap as it was in the late 1990s.

The cherry on top? Intel is teetering on the edge of being booted out of the Dow Jones Industrial Average, with AI chip champion Nvidia likely to take its place.

Layoffs, Cutbacks, and Cancellations

In early August, Intel’s CEO, Pat Gelsinger, announced a bold (and by bold, I mean drastic) plan to shrink the company’s workforce by 15,000 employees, cut its capital spending from over $25 billion to $20 billion, and slash its $3 billion dividend to… nothing.

“Our costs are too high, and our margins are too low,” Gelsinger lamented in a letter to employees.

Translation: We're in trouble.

Unsurprisingly, Intel’s stock price plummeted by another third in the days following the announcement. The Biden administration, keen on reviving domestic chipmaking, is probably getting a little nervous about their so-called "national champion."

Intel’s future now hangs on a new turnaround plan that Gelsinger is expected to present soon. Speculation is swirling that the plan might include even more layoffs, the sale of some non-core businesses, and possibly shelving the $32 billion factory planned in Germany.

But here’s the kicker: small changes won’t cut it.

To save Intel, Gelsinger will need to think bigger, bolder, and faster.

Gelsinger's Challenge: Fixing Decades of Missteps

But let us also give credit where it’s due—Gelsinger isn’t entirely to blame for Intel’s woes. He was brought back in 2021 to clean up the mess left by previous leaders who failed to see the tech industry's seismic shifts.

In the late 2000s, Intel was too busy basking in the fat profits of its PC business to notice the growing demand for mobile chips. While competitors were outsourcing production to fabless foundries like TSMC, Intel stubbornly stuck to making its own processors.

Manufacturing blunders delayed new chip launches, and AMD, a fabless competitor working with TSMC, swooped in to steal market share.

More recently, Intel missed the rise of AI chips entirely, leaving Nvidia to bask in its $3 trillion moment of glory—until the stock market corrected it a bit.

Gelsinger’s original plan was to split Intel into two: a fabless design unit and Intel Foundry Services (IFS). The idea was that without in-house production holding it back, the design unit could pick the best foundry for the job.

Meanwhile, IFS would win clients on its own merits, freed from its role as Intel’s captive manufacturer. Sounds good on paper, right?

Except for one problem: this plan assumed Intel could still rake in profits from CPUs and reclaim its lost leadership in Moore’s Law.

Spoiler alert—it hasn’t. One exasperated analyst even called the strategy “delusional.”

The Numbers Don't Lie—Intel's Decline Is Real

Intel’s revenues dropped from $79 billion in 2021 to just $55 billion by mid-2023, thanks to cooling demand for CPUs. And while there’s some promising progress in IFS, clients aren’t exactly lining up.

Case in point: Broadcom, a $700 billion chip designer, tested Intel’s new production process and concluded it wasn’t ready for prime time. Intel’s one big customer? Its own design unit, which, in a cruel twist, still outsources its high-end chips to TSMC.

Gelsinger remains optimistic, saying the new production process will be ready by next year, and Intel’s high-end chips will finally come home. He’s also betting on outside clients committing to $15 billion in loose purchase agreements with IFS.

Throw in cost savings, asset sales, $8.5 billion in grants, and $11 billion in loans from the U.S. government, and Gelsinger hopes to push Intel to the cutting edge. The grand plan? Eventually, IFS may even spin off as its own entity.

Radical Fixes or Risk Fading Into Obscurity

However, Intel’s situation is dire. Its once-healthy free cash flow of $10 billion per year turned deeply negative in 2022, and things have only worsened. Deals with private-asset managers to co-invest $10-15 billion in two factories have helped, but not by much.

A bolder leader might decide to ditch the AI chip race altogether—where Intel is embarrassingly behind—and sell the entire design unit to a fabless rival like Broadcom or Qualcomm. This would allow Intel to recapitalize its foundry without selling off assets piecemeal or angering the Germans by canceling that shiny new factory. We all know what happened the last time Germans were made angry.

No, we are not going to see WW3 if you were hoping for some frontline action.

Fabless chipmakers have a vested interest in Intel’s foundry success. Without it, they’re more dependent on TSMC, a risky proposition given Taiwan’s geopolitical tensions with China.

Intel’s collapse wouldn’t just be foolish from a business perspective—it could also be a national security risk. There’s even precedent for industry-wide cooperation: in 2012, ASML, a leading chipmaking equipment firm, raised funds from Intel, TSMC, and Samsung to develop next-gen tools. Intel’s $3.1 billion investment in ASML back then would be worth $50 billion today—almost 60% of Intel’s current value. Sold too early, right?

Potential backers, including tech giants like Amazon, Google, and Microsoft, could chip in to support Intel’s revival. These companies are designing more of their own AI chips, and Intel has expertise and top-notch engineers. While it may not deliver ASML-like returns, Intel might still be worth the bet.

In the end, Intel’s future isn’t about shrinking its way to success—though it certainly seems like that’s all it’s been doing lately.

To avoid fading into Silicon Valley obscurity, Intel will need bold, radical moves.

The question is, does Pat Gelsinger have the courage (and enough cash) to pull it off? Only time, and a few more earnings calls, will tell.

---