Intel's foundry business is gaining significant traction as three of the world's largest tech companies—Apple, Google, and Tesla—move to evaluate its advanced manufacturing capabilities. According to a report from CryptoComLearn, Apple is exploring Intel’s 18A-P node for its next-generation M-series processors, Google is investigating Intel’s EMIB packaging for the TPU v8e AI accelerator, and Tesla has confirmed it will use Intel’s 14A process for its Terafab AI computing complex in Austin, Texas—making Tesla Intel’s first external foundry customer.
Apple Eyes Intel 18A-P for M-Series Chips
Apple, historically reliant on TSMC for its A-series and M-series chips, is reportedly evaluating Intel’s 18A-P node. This enhanced version of the 18A process introduces PowerVia backside power delivery and RibbonFET gate-all-around transistors, offering competitive performance and power efficiency. If Apple commits, it would provide a major revenue boost to Intel’s foundry segment and signal a push toward supply chain diversification away from a single supplier.
Google Plans EMIB Packaging for TPU v8e
Google is exploring Intel’s embedded multi-die interconnect bridge (EMIB) packaging technology for its next-generation TPU v8e chip. EMIB enables high-bandwidth, low-latency connections between chiplets manufactured on different process nodes—a critical requirement for AI workloads. Previously, Google’s TPUs were designed by Broadcom and manufactured by TSMC. This evaluation marks a strategic shift toward alternative packaging suppliers to mitigate risk and accelerate time-to-market.
Tesla Becomes Intel’s First Public Foundry Customer
Tesla has confirmed it will deploy Intel’s 14A process at its Terafab AI complex in Austin. The facility will train neural networks for Tesla’s Full Self-Driving (FSD) system and Optimus humanoid robot. Tesla cited Intel’s domestic U.S. manufacturing capacity and closer geographic collaboration as key factors in choosing Intel over TSMC. This order is a symbolic win for Intel’s foundry business, providing a high-profile reference account.
Intel’s Foundry Losses Narrow, Yields Improve
Intel’s recent earnings report showed narrowing operating losses in its foundry segment, driven by improved process yields and growing customer interest. The company attributes this to a shift in CPU-to-GPU demand ratios fueled by the AI boom. Data center customers are increasingly looking for customized heterogeneous integration rather than monolithic high-performance CPUs, a trend that plays to Intel’s strengths in advanced packaging and flexible manufacturing.
Supply Chain Diversification Accelerates
Amid geopolitical tensions and recurring chip shortages, Apple, Google, and Tesla are actively seeking alternatives to TSMC. Intel’s “IDM 2.0” strategy—serving both internal and external customers—positions it as a credible second source. If the 18A and 14A nodes ramp on schedule, Intel could begin to challenge TSMC’s dominance in leading-edge logic in the second half of 2026. Analysts caution, however, that Intel still needs to prove its high-volume manufacturing capability and cost competitiveness to secure long-term commitments.