How to deal with Intel’s future raises fundamental concerns. One is: does Wall Street and the investment community really understand technology companies much at all?
There seems to be from most investors a broad understanding of new technologies like AI and quantum, but they view the tech too narrowly through quarterly earnings and forecasts that fail to foresee a 10-year, 20-year or even a 50-year future. It’s a Western mindset, almost tuned for the benefit of day traders who are willing to say, “My day came out green. What else do you want?” Meanwhile, the Chinese mindset (or more generally, an Asian mindset) looks a decade ahead, maybe even five decades ahead.
Since Intel started on its 15,000 layoffs course last fall and Pat Gelsinger was fired as CEO in December, tech reporters and analysts have been going crazy reporting all the various schemes for what to do next with Intel. These schemes include breaking off the chip design side from the fab side and tracking which company could or would be capable of investing in Intel, either partly or completely. The takeover names have included TSMC, Qualcomm, AMD, Broadcom and others.
The situation for Intel is lately complicated by a delay in the opening of two fabs in Ohio to the next decade and how the Trump administration will deal with tariffs on goods from Taiwan and TSMC, by far the world’s biggest producer of advanced chip nodes. When Trump met with various chip CEOs at Mar-a-Lago in mid-February, the result, based on analyst reports, was that TSMC could avoid potential tariffs by investing in Intel’s fab business (perhaps 20% of it or more), maybe even at the Ohio fabs, that would produce the latest chip nodes (even as advanced as 2nm) for Apple and Nvidia and others. This scheme could mean TSMC would need partners like Qualcomm or Broadcom.
Of course, none of this is official, and the various parties are not commenting.
Against that background, former Intel CEO and Chair Craig Barrett had the balls to come out and say in a Fortune piece on Friday that Intel should not sell its foundry business to TSMC and not break the company into two pieces—a design company and a foundry.
“In my opinion, a far better move might be to fire the Intel board and rehire Pat Gelsinger to finish the job he aptly handled over the past few years,” Barrett wrote.
His conclusion was audacious enough, but his reasoning gets back to the basic premise that many have expressed through the years, one holding that CEOs prior to Gelsinger and the boards overseeing them were financial types and not technology types with engineering chops--to Intel’s disadvantage. Every average business person knows the engineering mindset: engineers try things and often fail, usually with a view to what amazing new machines and processes can be produced and often with a view to generations into the future. It is not realistic enough for financial folks, working hard for investors who want to make their nut by the next quarter, if not today. (Well, more than a nut—20% margins at least.) This kind of critique is not leveled at whether Americans or oligarchs have the right to make more money; it is about leaders having a strong, lasting vision, whether wealthy or not.
Barrett wrote he was disappointed to “see people who ignore the intricacies of the semiconductor industry and proclaim simplistic solutions. It takes years to develop a new semiconductor manufacturing technology and ramp it into volume production.”
Noting that the “best technology wins,” Barrett argued that Intel’s leading technology is on par with TSMC’s 2nm technology and has a lead with the newest imaging technology known as high NA EUV lithography.
“Intel is about to regain its leadership [in semiconductor manufacturing], and the dumbest idea around is to stall that from happening by slicing the company into pieces,” Barrett wrote.
Multiple analysts agree with Barrett on not wanting to split up Intel.
“Intel is still better on its own rather than being acquired by others,” IDC analyst and long-time Intel observer Mario Morales said in an interview with Fierce. Any newcomer, including Qualcomm or Broadcom, will have to pay for the value of Intel and then also continue to invest. “That’s a big pill to swallow for any of these companies…I don’t believe TSMC should touch anything Intel has.”
Morales said the Intel board is apparently pushing for a breakup when it needs to be pushing for a new permanent CEO rather than continue much longer with two interim co-CEOs. “Pushing for a breakup distracts employees and investors,” he added. “If you wait this long and continue to cause chaos, you need a leader to bring together the entire company and almost shoot for a moonshot.”
“At the end of the day, the board is not thinking long term,” Morales added. “Value for investors is a short-term view, but this is a different thing-- for five to 10 years.”
No question, the future of Intel comes down to a conflict of values: Short-term gains over longer-term interests that matter to the broader tech community and US interests.