Intel has begun risk production of its enhanced 18A-P process node. The move, disclosed this week at the IEEE VLSI Symposium in Hawaii, signals fresh progress in the company’s bid to recapture ground in advanced manufacturing.
Officials say the node delivers 9 percent higher performance at the same power level as the base 18A. Or it cuts power consumption by 18 percent at equivalent performance. These gains come from coordinated changes in transistors, interconnects and design technology co-optimization. The improvements also bring better thermal characteristics and lower via resistance.
From Base Process to Performance Variant
The base 18A node already powers Intel’s Panther Lake Core Ultra Series 3 processors, launched in January. Yet even before that debut, the company had mapped out derivatives. 18A-P stands as the first. It stays fully compatible with existing 18A design rules. That compatibility matters. It lets chip designs shift over without major rework.
Analysts believe this feature could prompt Intel’s initial external foundry customers to jump directly to the enhanced version. Earlier plans had positioned the more aggressive 14A node as the primary offering for outsiders. CEO Lip-Bu Tan altered that approach. Now 18A-P looks positioned to serve as an early commercial vehicle for the foundry business. The Register first highlighted the shift and its implications for customer adoption.
Intel Foundry EVP Naga Chandrasekaran put it plainly. “Our updates and presentations at VLSI signal to Intel Foundry customers and partners that we are fully committed to leading edge process innovation over the long term.” His words, carried in Intel’s official release, underscore a message aimed squarely at potential partners. (Intel Newsroom, June 16, 2026)
But the path here was not smooth. Months earlier, Intel CFO David Zinsner admitted the company had overreached on 18A. “I would say it this way, I don’t know, early last year, I think the challenge around 18A was two things. One, we tried to do too much at once. And it took a while to get that settled. And I think second is, we were trying to play performance and yield and trying to improve both at the same time. It was like trying to fly the plane and fix the wing at the same time, basically.” Zinsner’s frank assessment, reported by The Register, captured the tension between ambition and execution.
So the arrival at risk production for 18A-P carries weight. Risk production means low-volume output to test and qualify the node. Hitting this stage on the schedule shared with customers last year offers tangible proof of stabilization. Intel claims the process now shows roughly 12 percent higher front-end frequency at iso-leakage. New RibbonFET options with mobility enhancements contribute. Power Boost dual contact transistors add more than 10 percent frequency at matched capacitance.
Thermal gains stand out too. The node cuts thermal resistivity by 20 to 40 percent. Via resistance on critical layers drops 10 to 30 percent. These changes matter for power-dense designs common in AI accelerators and high-performance computing. Sustained performance in tight thermal envelopes becomes more achievable. Tom’s Hardware detailed the thermal and interconnect upgrades in coverage published hours after the VLSI disclosures.
Intel also outlined 18A-PT, a variant aimed at 3D stacking with through-silicon vias. That version targets AI accelerator designers who want to pair logic with separately manufactured memory tiles. The broader 18A family, including these extensions, forms the foundation for Intel’s foundry pitch.
Market reaction was immediate. Shares of Intel rose in premarket trading Wednesday as word of the risk production milestone spread. Traders appeared to view the news as validation after years of questions about the company’s process leadership. Yet the competitive picture remains complex.
TSMC continues to dominate the pure-play foundry space. Intel must convince customers that its nodes match or exceed what the Taiwanese giant offers at similar schedules. Talks with Apple, reported in recent days, add intrigue. A potential manufacturing deal for some Apple silicon on 18A or 18A-P could provide high-profile validation. CNBC connected the production start directly to those discussions.
Additional details shared at VLSI point to longer-term bets. Engineers described work on complementary FET devices that stack NMOS and PMOS vertically for greater density. They also highlighted integration of gallium nitride power devices with silicon logic. Such steps aim to combine control circuitry and high-power transistors on one die. These remain under development. They illustrate Intel’s stated intent to sustain innovation beyond the current 18A platform.
The stakes sit high. Intel has poured billions into its process roadmap. Success with 18A and its derivatives could restore confidence in its manufacturing prowess and attract the external customers needed to fill fabs. Failure would reinforce doubts about its ability to compete at the leading edge.
Production of base 18A already runs in U.S. facilities in Arizona and Oregon. Those sites represent some of the most advanced semiconductor plants globally, according to company executives. High-volume manufacturing of 18A is slated for later in 2025, with 18A-P expected to follow as demand materializes.
Customers will watch yields closely. Past challenges with 18A made clear that performance claims must translate into reliable, high-yielding production. The risk-production milestone offers an early data point. Full qualification and volume ramp will provide the definitive test.
And the timing aligns with surging demand for AI-capable processors. Both internal Intel products and potential foundry wins stand to benefit if the nodes deliver on promised efficiency. 18A-P’s focus on performance and thermal headroom positions it well for exactly those workloads.
Intel has spent the past year refining its foundry strategy under new leadership. The emphasis now falls on execution and clear communication of progress. This week’s VLSI updates fit that pattern. They deliver specific metrics, compatibility assurances and a roadmap extension all at once.
Whether that proves enough to sway major external designers remains to be seen. Yet the company has moved past the admission of earlier missteps. It now points to silicon in limited production and a set of measurable gains. For an industry that rewards results over promises, the shift carries significance.
Further enhancements to the 18A family are already in planning. The platform, Intel insists, will see continued refinement for years. That long-term commitment forms a core part of the pitch to foundry customers who need predictable innovation cycles.
The coming months will reveal how quickly 18A-P advances from risk production to broader availability. They will also show which partners choose to board. For now, Intel can claim it has met its publicly shared timeline. In a business where delays have become costly, that counts as tangible forward motion.
WebProNews is an iEntry Publication