Intel’s Custom ASIC Revenue Climbs 50% to $1B Annual Run Rate

INTCINTC

Intel’s custom ASIC segment grew over 50% in 2025, hitting an annualized $1 billion revenue run rate after forming a dedicated Central Engineering Group. The business addresses a $100 billion total addressable market by bundling design services, IP building blocks, and in-house manufacturing.

1. Violent Q4 Sell-Off Belies Underlying Strength

In the days following Intel’s fourth-quarter earnings release, the stock plunged over 20% as investors reacted to cautious guidance for the first quarter of fiscal 2026. While revenue of $14.8 billion and non-GAAP earnings per share of $0.76 narrowly exceeded consensus, the company warned of supply constraints in its leading nodes that would depress near-term output. Despite the sell-off, gross margin improved sequentially by 120 basis points to 36.4%, reflecting better mix in data-center and AI products. Management reiterated full-year capital expenditures of $20 billion and highlighted a continued path to positive free cash flow in the second half, underpinned by working-capital discipline and yield improvements across its Intel 7 and Intel 3 process technologies.

2. Emerging High-Margin Businesses Powering the Turnaround

Intel’s custom ASIC and AI PC initiatives are gaining traction and offer a compelling counterpoint to its near-term manufacturing bottlenecks. In 2025, the custom ASIC segment grew over 50% year-over-year to exceed a $1 billion annualized run rate, driven by demand for networking and AI acceleration chips designed on Intel’s 18 ångström platform. Meanwhile, the newly launched Core Ultra Series 3 processor line now powers more than 200 distinct laptop and desktop designs, targeting the fast-growing AI-enabled PC market, which IDC projects will reach $25 billion by year-end. These businesses carry gross margins north of 50% and are expected to contribute up to 15% of overall revenues by mid-2026, bolstering the company’s longer-term profitability profile.

3. Foundry’s Make-or-Break Moment in 2026

After years of capital investment in leading-edge fabs, Intel’s foundry unit faces a critical test: securing external customer commitments before ramping its next-generation Intel 14A node. Management has explicitly tied capacity build-outs to signed supply agreements, a departure from its prior “build-it-and-they-will-come” approach. Risk-production tooling for 14A is slated to begin in late 2026, with volume output expected in 2028—contingent on firm customer orders in the back half of ’26. Given global foundry capacity shortages at competitors and the company’s unrivaled packaging capabilities, Intel believes it can win marquee cloud and AI accelerator clients this year. Success would not only validate its foundry strategy but also drive multi-billion-dollar revenue streams and meaningfully de-lever the capital-intensive nature of its manufacturing investments.

Sources

FSFFS
+14 more