Intel CEO Admits ‘Bankrupt’ Narrative Deterred Top Talent During Turnaround Push
Table of Contents
The struggle to recruit amidst a crisis
When Lip-Bu Tan stepped into the CEO role at Intel, he didn’t just inherit a struggling product line or a lagging manufacturing process; he inherited a perception problem. Speaking at the JP Morgan Global Technology, Media and Communications Conference on Tuesday, Tan candidly shared that the company’s financial instability had become a deterrent for the very talent needed to fix it.
According to Tan, potential recruits were not just hesitant—they were blunt. “I tried to recruit some talent. They said, ‘It’s almost a bankrupt company, why should I join you?’” Tan recalled. For a company that once defined the gold standard of Silicon Valley engineering, the admission that top-tier talent viewed the firm as a sinking ship underscores the depth of the crisis Intel has faced over the last several years.
Stabilizing the balance sheet became the immediate priority. This was achieved through a combination of government intervention and strategic partnerships. Tan noted that securing equity investment from the Trump administration—via the conversion of CHIPS program funds—provided a critical safety net. Further credibility arrived via personal industry ties, including a $5 billion commitment from Nvidia CEO Jensen Huang and backing from Softbank’s Masayoshi Son, a former Intel board member.
“So far, knock on wood, I made money for them, and they’re quite happy,” Tan said, noting that the improved financial standing eventually allowed Intel to buy back a stake from Apollo, reducing earnings-per-share dilution.
Beyond the 18A node: A long-term roadmap
With the immediate existential threat dampened, Tan is now pivoting the conversation toward Intel’s long-term viability as a foundry. While much of the industry focus has been on the current 18A node, Tan is already signaling a leap toward 10A and 7A technologies.
The 18A process is currently showing a 7 percent monthly yield improvement, and Tan describes the next-generation 14A node as being “ahead of schedule” relative to year-end targets. However, the real goal is to convince third-party customers that Intel isn’t just a one-hit wonder with a single node. “People don’t go to you just for one node. They’re looking for the roadmap for the future,” Tan explained.
The timeline for these advancements is aggressive but distant. Tan indicated that risk production for 14A is slated for 2028, with volume production in 2029—putting it on a similar trajectory to TSMC’s A14. This suggests that the elusive 10A and 7A nodes, which could push Intel into sub-nanometer territory, are unlikely to see commercial application before 2030.
The Foundry Pivot
Intel’s transition into a foundry service for third-party clients is a fundamental shift for a company that historically kept its manufacturing proprietary. To bridge the expertise gap, Tan recently poached Shawn Han, a 30-year Samsung Foundry veteran, as SVP of Foundry Services.
The appetite for Intel’s services seems to be growing, particularly for the 18AP node—a performance-focused variant of 18A. Tan claimed that some customers are so eager to secure capacity that they are offering to prepay for wafer substrate materials. “If you are serious to use our EMIB-T [packaging technology], can you help me on the substrate prepay? They jump on it,” Tan said, adding that these commitments involve billions of dollars over the next few years.
Betting on Agentic AI
The overarching gamble for Intel’s recovery rests on a shift in AI architecture. While GPUs have dominated the training phase of AI, Tan believes the era of “agentic AI” and inference workloads will swing the pendulum back toward CPUs.
In the early stages of the AI boom, the ratio of CPUs to GPUs for training was roughly 1:8. However, Tan reports that startups are now seeing a much higher demand for CPU power to handle autonomous agents and inference. “I can start to see… customers have said to me, more like 1:1. And now even some of them tell me it’s 4:1,” Tan said, referring to the CPU-to-GPU ratio.
Despite the optimistic outlook on AI and manufacturing, Intel’s financial recovery remains a work in progress. The company reported a loss of $267 million on revenues of $52.9 billion during 2025, a narrowing from the massive $18.8 billion loss the previous year, but one that leaves the company still fighting to return to consistent profitability.
