Intel CEO Pat Gelsinger’s abrupt and immediate “retirement” from the chip giant ends what was a tumultuous time for the former VMware head who was brought back to Intel in hopes of steering a turnaround. It also further calls into question Intel’s ability to make headway into the rapidly expanding artificial intelligence (AI)-fueled data center space.

Gelsinger left Intel less than four years after re-joining the semiconductor firm under the hallmark of turning around what had been a somewhat aimless operation. Gelsinger had been something of a legend within Intel’s halls, having spent nearly three decades at the company where he served as CTO, helped architect the 80486 processor, led 14 different microprocessor programs, and played key roles in the Core and Xeon families.

Gelsinger first left Intel in 2009 to take over as president and COO of Dell EMC and later spent eight years as CEO of VMware.

Gelsinger did make some progress toward guiding Intel’s turnaround. This was highlighted by Gartner, which reported Intel had reclaimed the top spot in worldwide semiconductor revenues in 2023. Gelsinger was never able to get the ship righted.

However, financial struggles continued.

Intel recently announced drastic cost cutting measures revolving around plans to slash approximately 15,000 jobs by the end of this year, which is part of a plan to cut $10 billion in spending. Intel is also in the process of moving its Edge business into its Client Computing Group (CCG) and refocusing its Network Edge (NEX) operations on networking and telecommunications.

Gelsinger told investors during Intel’s most recent earnings call that part of that internal reorganization will see it “re-establishing product portfolio leadership by narrowing our focus on fewer projects with the top priority being to maximize the value of our x86 franchise across the client, edge, and data center markets.”

Some progress toward that goal was highlighted by Intel scoring a custom chip deal to power data centers for Amazon Web Services (AWS).

That deal calls for Intel to produce an artificial intelligence (AI)-focused chip using its 18A processing node architecture and custom Xeon 6 chips on its Intel 3 architecture. More importantly, those chips are scheduled to be produced at an Intel facility in New Albany, Ohio.

“Solid progress on it,” Gelsinger said of those plans. “Clearly, next year, there’s not a lot of financial benefit from it because we’re only ramping late in the year. Thus, we’ll be giving more qualitative metrics on progress as we go through the year.”

Those efforts are targeted at a surging AI-fueled data center opportunity that is pushing hyperscalers like AWS, Microsoft Azure, and Google Cloud Platform (GCP) to invest heavily into their data center infrastructure.

“New AI-oriented services and technology are helping the major cloud providers to ride a wave – new capabilities lead to increased demand, which leads to increased revenues, which then enables more investment in underlying technologies,” Synergy Research Group (SRG) Chief Analyst John Dinsdale wrote in a recent report.

The AWS deal provides Intel with a significant boost toward garnering a larger share of this opportunity, which continues to be dominated by rivals like Nvidia, AMD, and Arm.

Gelsinger specifically pointed to some enterprise-focused AI efforts where it might find further traction with Gaudi 2 AI Accelerators.

“As you go into enterprise AI, we expect to place a more prominent role,” Gelsinger said. “Databases, embedding, refinement are much more attuned to CPU workloads and our strategy there is CPU plus accelerator or CPU plus Gaudi. So we see the enterprise use cases having a very long life associated with them going forward.”

Despite those wins, Dave Novosel, who is a financial analyst with corporate bond analyst firm GimmeCredit, noted Intel’s foundry ambitions were burning through its cash reserves.

“Foundry losses exceeded $5 billion last year and are likely to exceed $13 billion this year. External chip sales have been minimal,” Novosel wrote in a report. “Meanwhile, Nvidia has taken a commanding lead in the manufacturing of AI chips, leaving Intel well behind.”

What’s next for Intel’s data center plans? Gelsinger's departure also came just days after Intel secured nearly $8 billion in government aid to set up a chip foundry in the United States. Those plans are tied to hiring up to 10,000 new permanent jobs at those facilities and to help the U.S. ween itself from chips made in China.

Alvin Nguyen, senior analyst at Forrester Research, explained in a blog post that “tech execs need to pay attention to who Intel brings in as CEO. What the company does next will impact your purchasing decisions for your workplace, data center, and cloud services.”

Nguyen noted that data center focus is increasingly important as Intel rivals continue to show robust growth, especially in terms of AI development.

“Data center product growth, especially in AI with Nvidia and now AMD, shows Intel falling behind in a rapidly growing market,” Nguyen wrote. “Its data center CPUs are also behind AMD’s offerings, but the new Xeon 6 release following AMD’s latest processors is an opportunity to gain back some thought leadership and market share.”

Intel did recently sign an agreement with AMD to form the x86 Advisory Group, which is focused on steering the future of that widely adopted platform. Other members of that founding group include Broadcom, Dell Technologies, Google Cloud, Hewlett Packard Enterprise (HPE), Microsoft, Lenovo, Oracle, and Red Hat.

“We definitely want to be very front footed with x86 for a full range of use cases, but also the AI use cases as well, and the industry is quite interested in joining us, participating and expanding the world’s greatest architecture of all time,” Gelsinger said of those efforts. “The most industry influence, the broadest number of ISVs and applications and continuing that momentum forward.”

That influence will now be under new leadership.