Stories How Intel Is Rebuilding Around 18A and Foundry

How Intel Is Rebuilding Around 18A and Foundry

Intel's current manufacturing strategy is an attempt to solve two problems at once.

restore manufacturing leadershipturn manufacturing into an external business

Either goal would be difficult alone.

Intel is attempting both.

Step 1 — Accelerate the Process Roadmap

Intel pursued an aggressive sequence of process nodes intended to restore competitiveness.

The roadmap culminated in 18A entering production in 2025. [S16]

Step 2 — Introduce RibbonFET and PowerVia

18A combines two major technology changes:

  • gate-all-around RibbonFET transistors;
  • backside power delivery through PowerVia.

This makes 18A more than a simple optical shrink.

Step 3 — Put Internal Products on the Node

Panther Lake provides client volume.

Clearwater Forest provides server volume.

Internal products create factory learning before external foundry volume becomes large.

Step 4 — Build a Foundry Ecosystem

External customers need more than wafers.

Intel has worked with:

  • EDA suppliers;
  • IP providers;
  • packaging partners;
  • design-service companies.

The goal is to make an Intel process usable by companies that did not design chips inside Intel. [S34]

Step 5 — Sell Packaging as a Separate Advantage

Advanced packaging can attract customers before they move a full leading-edge design to Intel.

That gives Intel an intermediate commercialization path.

Step 6 — Change Internal Accounting

Intel now reports Products and Foundry more like separate companies.

The purpose is to expose:

  • manufacturing cost;
  • transfer pricing;
  • utilization;
  • foundry losses;
  • product economics.

This is operationally uncomfortable but strategically useful.

Step 7 — Add External Customers

This is the hardest step.

Intel must convince chip designers to risk major product generations on Intel manufacturing.

A leading-edge foundry relationship can last years and involve billions of dollars of customer product revenue.

Trust is therefore slow to build.

Step 8 — Make 14A Demand-Led

Under Lip-Bu Tan, Intel has said future 14A investment will depend on confirmed customer commitments. [S6]

This changes the sequence from:

build technology
then hope customers arrive

toward:

co-design technology with lead customers
then scale investment with demand

U.S. Strategic Support

Government funding and defense programs reduce some financing burden and support domestic capacity. [S29][S36]

But government support does not solve commercial competitiveness.

A foundry still needs customers, yield and cost discipline.

The Core Tension

Intel Products wants the best manufacturing for Intel products.

Intel Foundry wants to be neutral and customer-first.

The company must make both true simultaneously.

What Success Would Mean

A successful rebuild would create:

competitive Intel CPUs
+
external foundry customers
+
high fab utilization
+
advanced packaging leadership
+
repeatable process roadmap

That would restore the logic of vertical integration in a modern, more open form.

What Failure Would Mean

If external customers remain small and manufacturing economics stay weak, Intel could carry the cost of a global leading-edge fab network without enough external volume to justify it.

That is why Foundry is both Intel's biggest strategic opportunity and one of its biggest financial risks.