Musk Visits Intel Oregon Fab to Scout 18A for xAI Chips

Elon Musk personally visited Intel's advanced semiconductor fabrication facility in Oregon this week, touring the company's cutting-edge wafer production lines as he seeks foundry capacity for xAI's next-generation artificial intelligence chips. Musk confirmed the visit himself on social media, sharing a photo alongside Intel CEO Lip-Bu Tan, signaling the deepening strategic relationship between the two tech titans.

The Oregon facility operates under Intel Foundry Services (IFS), the chipmaking giant's contract manufacturing division, and currently produces advanced chips — including the upcoming Panther Lake processors — using Intel's 18A process technology. The visit underscores Musk's hands-on approach to securing the semiconductor supply chain his AI ambitions demand.

Key Takeaways From Musk's Intel Visit

• Musk toured Intel's Oregon fab, which houses 18A process node production lines
• Intel CEO Lip-Bu Tan personally hosted the visit, highlighting its strategic importance
• Intel and xAI have already agreed on 2 major collaborations, including the Terafab project
• xAI's next-gen AI chips will reportedly use Intel's even more advanced 14A process technology
• Intel is positioning both 18A and 14A as 'backbone' nodes for foundry customers
• The visit likely aimed to evaluate production readiness and capacity constraints ahead of mass production

Two Major Deals Already in Place Between Intel and xAI

Musk's factory tour was far from a casual sightseeing trip. Intel and Musk's companies have already formalized 2 significant partnership agreements that could reshape the competitive dynamics of AI chip manufacturing.

The first collaboration centers on the Terafab project, a massive semiconductor manufacturing initiative that Musk has publicly championed. The second — and arguably more consequential — deal involves Intel manufacturing xAI's next-generation AI processors using the company's 14A process node, which represents Intel's most advanced chipmaking technology currently in development.

Musk has previously stated publicly that the Terafab project will leverage Intel's 14A process technology. This positions Intel as a direct competitor to TSMC, the Taiwanese foundry giant that currently dominates advanced chip manufacturing for companies like Nvidia, Apple, and AMD.

For Intel, landing xAI as a major foundry client represents a critical validation of its turnaround strategy under CEO Lip-Bu Tan. The company has been aggressively investing billions of dollars to rebuild its manufacturing competitiveness after years of falling behind TSMC and Samsung in process technology leadership.

Why 18A Matters: Intel's Bid to Reclaim Manufacturing Leadership

The Intel 18A process node represents the cornerstone of Intel's semiconductor manufacturing renaissance. It introduces several breakthrough technologies that Intel hopes will leapfrog the competition and attract high-value foundry customers.

Key technical innovations in 18A include:

• RibbonFET — Intel's implementation of gate-all-around (GAA) transistor architecture, replacing the aging FinFET design that has dominated chip manufacturing for over a decade
• PowerVia — a backside power delivery network that separates power routing from signal routing, improving performance and power efficiency
• Significant density improvements over the previous Intel 20A node
• Competitive positioning against TSMC's N2 (2nm) process node expected in similar timeframes

Panther Lake processors will be the first products manufactured on 18A, serving as a critical proof point for the technology's maturity. For potential foundry customers like xAI, the performance of these initial products will be a key indicator of whether Intel can deliver on its ambitious promises.

The fact that Musk chose to visit the fab in person — rather than simply relying on engineering reports — suggests he wants firsthand visibility into the production environment, yield rates, and potential bottlenecks before committing to large-scale orders.

The Strategic Context: Why Musk Needs His Own Chip Supply Chain

Musk's aggressive pursuit of foundry partnerships reflects a broader strategic imperative. His AI company xAI, which operates the Grok large language model and runs one of the world's largest AI training clusters in Memphis, Tennessee, has an insatiable appetite for compute power.

Currently, the AI chip market is overwhelmingly dominated by Nvidia, whose H100 and H200 GPUs power the vast majority of AI training workloads globally. Nvidia's chips are manufactured exclusively by TSMC, creating a supply chain concentration risk that Musk — and many other tech leaders — view as strategically untenable.

By developing custom AI chips manufactured by Intel, Musk could achieve several objectives simultaneously. He would reduce dependence on Nvidia's increasingly expensive and supply-constrained GPUs. He would gain chips optimized specifically for xAI's unique workloads and architectures. And he would diversify his supply chain away from TSMC, which faces geopolitical risks given its location in Taiwan.

This strategy mirrors moves by other tech giants. Google designs its own TPU chips, Amazon has developed Trainium and Inferentia processors, and Microsoft recently unveiled its Maia AI accelerator. Custom silicon is rapidly becoming a competitive necessity rather than a luxury for companies operating at AI scale.

Intel's Foundry Ambitions Get a Major Credibility Boost

For Intel, Musk's visit and the xAI partnerships represent perhaps the most significant external validation of Intel Foundry Services since its creation. The division has struggled to attract major customers, with skeptics questioning whether Intel could successfully operate as both a chip designer and a contract manufacturer for competitors.

CEO Lip-Bu Tan, who took the helm earlier this year, has made revitalizing Intel's foundry business a top priority. Landing xAI as a marquee customer sends a powerful signal to the broader semiconductor industry that Intel's manufacturing technology is ready for prime time.

During Intel's recent earnings season, company executives revealed that both 14A and 18A process nodes are being offered to foundry customers as technology 'backbone' options — meaning they are positioned as stable, long-term manufacturing platforms rather than transitional nodes. This is significant because foundry customers need confidence that a process technology will be supported and optimized for years, not abandoned after a single product generation.

The xAI deal also helps Intel justify its enormous capital expenditure program. The company has committed tens of billions of dollars to new and upgraded fabrication facilities across the United States and Europe, supported in part by CHIPS Act subsidies from the U.S. government. Having a high-profile customer like xAI helps demonstrate return on that investment to shareholders and policymakers alike.

What This Means for the AI Chip Landscape

Musk's Intel partnership could trigger a significant shift in how AI chips are sourced and manufactured. If xAI successfully produces competitive custom silicon on Intel's 14A process, it could encourage other AI companies to explore similar arrangements — potentially breaking TSMC's near-monopoly on cutting-edge AI chip manufacturing.

For developers and businesses building on AI infrastructure, the implications are substantial. More diverse chip manufacturing means potentially shorter wait times for AI hardware, more competitive pricing, and greater supply chain resilience. It also means the AI hardware ecosystem could become more fragmented, with different chips optimized for different workloads.

The geopolitical dimension is equally important. U.S. policymakers have been pushing aggressively to reshore semiconductor manufacturing, and a high-profile partnership between America's most prominent tech entrepreneur and its largest chipmaker perfectly aligns with that national security agenda. The Terafab project, in particular, could become a flagship example of domestic semiconductor self-sufficiency.

Looking Ahead: Timeline and Next Steps

Several critical milestones will determine whether this partnership delivers on its promise. Intel's 18A process must first prove itself in volume production with Panther Lake processors, likely in the second half of 2025. The 14A node, which xAI plans to use for its custom AI chips, is further out on the roadmap.

Key developments to watch include:

• Panther Lake yield rates — strong yields on 18A would boost confidence in Intel's manufacturing capabilities
• 14A development timeline — any delays could push back xAI chip production
• Terafab construction progress — the massive fab project requires years of construction before reaching production
• xAI chip architecture details — what specific AI workloads the custom chips will target
• Competitive response from TSMC and Nvidia — both companies are unlikely to stand still

Musk's personal involvement in evaluating Intel's manufacturing capabilities signals that xAI's custom chip program is a top priority, not a side project. If Intel can execute on its technology roadmap, the partnership could mark a turning point in both companies' trajectories — and in the broader AI semiconductor landscape.

The stakes could hardly be higher. For Intel, this is a chance to prove that its foundry business can compete at the highest level. For Musk, it is about building the vertically integrated AI infrastructure needed to keep xAI competitive against deep-pocketed rivals like Google, Microsoft, and OpenAI. And for the industry, it represents a potential rebalancing of semiconductor manufacturing power that has tilted heavily toward East Asia for the past 2 decades.