INTC 3Q25 CC: "advanced packaging activities continue to progress well"
Lastly, our advanced packaging activities continue to progress well, especially in the area like EMIB and EMIB-T, which we have 2 differentiations. Like our Intel products, my conviction in the market potential for Intel Foundry continue to grow.
The rapid expansion of critical AI infrastructure is fueling unprecedented demand for wafer capacity and advanced packaging services that present a substantial opportunity demanding multiple suppliers. Intel Foundry is uniquely positioned to capitalize on this unprecedented demand as we execute. As I mentioned last quarter, our investment in foundry will be disciplined, and we will focus on capability and scalability, giving us flexibility to ramp quickly, and we will only add capacity when we have committed external demand. Building a world-class foundry is a long-term effort founded on trust. As a foundry, we need to ensure that our process can be easily used by a variety of customers, each with a unique way of building their own products.
Copilot:
Intel was a leader in glass substrate R&D for advanced packaging, but under Lip-Bu Tan, it’s pivoting to external suppliers—still aiming to be a major player by leveraging its hybrid bonding and BEOL strengths.
Here’s the full picture:
Intel’s Packaging Strengths
- Advanced packaging is a core Intel capability, especially in hybrid bonding, EMIB, and Foveros 3D stacking.
- Intel is one of the few firms with full BEOL (back-end-of-line) integration, crucial for AI accelerators and high-power SoCs.
- Lip-Bu Tan called packaging a “substantial opportunity” during recent earnings calls, signaling strategic focus.
Glass Substrate R&D: What Happened
- Intel spent years developing glass substrates for next-gen packaging—offering better thermal stability, flatness, and via density than organic substrates.
- Tan recently shifted strategy, ending in-house glass R&D to source from external vendors like JNTC, SK hynix (Absolics), and possibly Samsung.
- This move cuts cost and accelerates integration, allowing Intel to focus on 14A/18A nodes and foundry viability.
Strategic Partnerships
- Samsung may invest in Intel’s packaging business, potentially forming a joint venture to compete with TSMC’s CoWoS dominance.
- Intel is reportedly licensing its glass substrate IP, monetizing past R&D while streamlining operations.
Competitive Position
- TSMC leads in volume and CoWoS adoption, but Intel’s vertical integration and hybrid bonding give it an edge in certain AI and HPC segments.
- Glass packaging is heating up: JNTC expects $735M revenue by 2028, and SK hynix is scaling its pilot lines.
- Intel’s pivot suggests it wants to stay agile, not own the substrate supply chain.
Your read is sharp
Intel was deep in glass R&D and still sees packaging as a strategic lever. Tan’s shift to external sourcing doesn’t weaken Intel’s position—it may actually strengthen its execution by reducing capex and focusing on integration.
Want to riff on how this affects ATP flow, substrate demand in Southeast Asia, or how glass packaging parallels pellicle design in EUV? Or how Intel’s packaging pivot might reshape AI accelerator competitiveness vs. NVIDIA and AMD? |
