And now a quick word from my partner Dylan Byers on Kimmelgate…

There was a time when Intel didn’t need help from anyone. The “Intel Inside” marketing campaign, launched in 1991, was genius. It encouraged consumers who were buying branded P.C.s—Dell, HP, IBM—to demand their computers also have the fastest central processing units (or C.P.U.s) in them. And those C.P.U.s were made by Intel. If you bought into “Moore’s law,” which essentially predicts that computing power should double every two years, Intel seemed unstoppable.

Moore’s law was named for Gordon Moore, the co-founder of Intel and its C.E.O. from 1979 to 1987. Alas, the leaders who followed Moore and the legendary Andy Grove, including Paul Otellini, Brian Krzanich, Bob Swan, and Pat Gelsinger were never really up to the task, leading to years of market misjudgments, leadership snafus, and general arrogance.

Lip-Bu Tan was supposed to be better. Born in Malaysia and raised in Singapore before eventually moving to California, Tan is the founder of venture capital firm Walden International, which has numerous investments across Asia, including in SMIC, or Semiconductor Manufacturing International Corporation, the Chinese rival to TSMC—which, of course, is in Taiwan. From 2009 to 2021, Tan was C.E.O. of Cadence Design Systems, a leader in the semiconductor software design industry. In March, he was appointed C.E.O. of Intel, where his performance will now be judged, for a while at least, based on the recent, strange turn of events and his ability to manage, and satisfy, this interesting collection of new shareholders.

To get an understanding of precisely how Intel had become almost an afterthought in the semiconductor industry, and what the new $16 billion infusion will mean for the company, I rang up Stacy Rasgon, a managing director and senior analyst at Bernstein Research. Rasgon, who covers the semiconductor industry, has a Ph.D. in chemical engineering from MIT and has been on Extel’s (f.k.a. Institutional Investor) All-America Research Team every year since 2010, with several appearances as the top-ranked research analyst.

“Intel has been a total disaster,” he told me. During the heyday of the P.C., Intel had “legitimate, monopoly-sized economics,” but then it got “fat” and “dumb.” For example, around 2014, Intel’s biggest competitor was AMD, which had a 0.1 percent share of the chip market; Intel had a 99.9 percent share. Intel management was dismissive of AMD, Rasgon said, “and they got lazy.”

Alas, Moore’s law didn’t hold as the chips got smaller and smaller, and Intel whiffed on a bunch of new market opportunities, Rasgon told me, including chips for mobile devices. Management didn’t think the volumes would be sufficient to overcome Intel’s high-cost model, Rasgon said. But as it turned out, the volume of smartphone sales would be 100x Intel’s projections.

When the company tried to catch up in the mobile market, it lost billions along the way. “I never thought it would get this bad,” Rasgon said, noting that at least Tan wasn’t “delusionally optimistic” like at least one of his predecessors. Tragically, Intel also downplayed the G.P.U. (graphics processing unit) opportunity that Nvidia capitalized on for A.I. “They were dismissive, I think, at first of the [A.I.] trend,” Rasgon said. “And then when they realized it was going to be a real problem, they decided to try to acquire [the capacity to make A.I. chips].”

But Intel’s M&A track record “is absolutely horrendous,” Rasgon noted, citing flops such as deals for Altera ($16.7 billion); Mobileye (another $15 billion); DSP (“sold for peanuts” and a $300 million loss); McAfee (“To this day I don’t know why they bought that”); and an A.I. startup called Nervana Systems, which Intel “eventually wound up shutting it down.”

Regarding the Trump administration’s conversion of the $9 billion CHIPS grant into Intel equity last month, Rasgon wrote in an August 25 note to investors that “the valuation seems very low” in exchange for money the company was supposed to get for free. Nevertheless, he wrote, “Fans of the CHIPS-for-equity deal have pointed out that governmental support is nothing new in semis, and have deemed it welcome to have Trump as an engaged stakeholder, while dissidents have decried the idea of governmental meddling in the free market (up to and including making allusions to communism).” As for Nvidia’s $5 billion investment, he concluded that it “seems like money well spent at this point,” with considerable understatement, pointing out that its stake in Intel had increased in value by $1 billion on the news alone.

But Rasgon has still been left scratching his head. Intel now has some $16 billion of new cash, but, as he told me, it really doesn’t need it. While the money could help Intel expand its manufacturing capacity, what the company really needs is more people buying its products. “They’ve got no customers,” he said. “And so there’s been a lot of speculation that, hey, the next thing the government will do is force customers to use them. Personally, I think that is ridiculous. I don’t think you can. I don’t think you need to. If Intel could prove that they can make products in high volume that meet spec—and which have a good cost structure and are available on time—they will have customers lined up around the block.”

Semiconductors are “the most complicated things that humanity has ever built,” Rasgon told me, and over the years, Intel has lost the capability to deliver what customers want on a sustained basis. “To get customers,” he said, “you have to prove you can deliver. And can they deliver? Who knows? Call me in three years.”

Even with the new money, the seeming votes of confidence from SoftBank and Nvidia—and the boost, or whatever it is, from Trump—Intel remains “fundamentally in a rough spot,” Rasgon said. But he wants Intel to succeed.  “I do think it’s strategically important to have a successful [domestic] leading-edge semiconductor company,” he said, “but they’ve got to do it themselves.” There is no “secret sauce” that Nvidia’s Jensen Huang can share, despite the $5 billion investment. “It’s thousands and thousands of engineers working long, long hours for years.”

All the new cash from, in part, two fairly sophisticated investors, has driven Intel’s stock up in the near term, which is nice, I suppose, but it’s unlikely to solve a problem more than a decade in the making. “Trump’s not going to be down in the fucking lab trying to improve the yields on the chips,” Rasgon said.

As far as any recommendation for investors about whether to start loading up on Intel, Rasgon told me he’d short the stock if it weren’t so volatile and now that it’s become a bit of a meme stock. “You get a headline like this and it goes up 30 percent in your face,” he said. “How can you short it?” (As always, this is not investment advice.) But Rasgon doesn’t want to own the stock, either. “I’ve been preaching avoidance,” he said. “It’s an event stock. But you don’t know what the event is.”

Need help? Review our FAQ page or contact us for assistance. For brand partnerships, email ads@puck.news.

You received this email because you signed up to receive emails from Puck, or as part of your Puck account associated with {{customer.email}}. To stop receiving this newsletter and/or manage all your email preferences, click here.

Puck is published by Heat Media LLC. 107 Greenwich St, New York, NY 10006