[News] Intel CFO Says 14A Defect Reduction Fastest Since 22nm; Customer Talks Reportedly Grow Ahead of 2028 Ramp

Intel’s latest 14A progress is drawing attention after CFO David Zinsner said the node’s defect density is improving faster than expected. According to a transcript from Investing.com, Zinsner said at Deutsche Bank’s 2026 Technology Conference that 14A defect density was tracking better than Intel’s target curve, with the company seeing its fastest pace of defect reduction since 22nm, which he described as one of Intel’s best nodes. Still, Tom’s Hardware notes several caveats. Intel’s comments do not mean 14A’s current defect density matches that of 22nm at a comparable stage before mass production. Advances in wafer processing and inspection also mean what Intel counts as a defect today may differ from 16 years ago. More importantly, defect density alone does not directly translate into product yield.

As ijiwei notes, the defect-density curve is therefore a leading indicator rather than a definitive measure of success. The real tests will come with risk production. Whether Intel can meet yield targets, bring costs under control, and deliver planned capacity will ultimately determine whether 14A lives up to the “best since 22nm” comparison.

Still, ijiwei notes that 14A introduces three new technologies at once: second-generation RibbonFET, PowerDirect backside power delivery, and High-NA EUV. Each presents an industry-level challenge, making 14A significantly more complex than 22nm and even 18A. Against this higher level of complexity, the faster decline in defect density suggests a genuine improvement in Intel’s learning-curve management.

Customer Interest Emerges as Key 14A Signal

Beyond the technical progress, ijiwei highlights customer sentiment as perhaps the more significant signal from Zinsner’s remarks. Defect density is an internal metric, while external orders will ultimately determine the success of Intel’s foundry business.

Zinsner’s comments last week offered some encouraging signs on that front. According to the Investing.com transcript, Intel’s internal product teams are now designing products on 14A. More importantly, Zinsner said engagement with external foundry customers has “significantly increased.” CEO Lip-Bu Tan and his team are now meeting customers weekly, with discussions shifting from reviewing technical data toward questions about available capacity and supply. Zinsner added that Intel plans to release the 14A 0.9 PDK in October, begin risk production in the second half of 2027, and ramp into high-volume production in 2028.

That timeline could also put Intel on a closer footing with TSMC at the leading edge. According to TSMC, N2 entered mass production at the end of 2025, while A16 is scheduled for mass production in the second half of 2026, A14 in 2028, and A13 in 2029. ijiwei notes that if Intel brings 14A into high-volume production as planned in 2028, its rollout would coincide with TSMC’s planned A14 production, putting the two companies on unusually close leading-edge timelines after more than a decade of Intel playing catch-up.

Read more

• [News] Intel’s Advanced Packaging, Led by EMIB-T, Set to Hit Its Stride in 2029; DRAM Manufacturing Ruled Out
• [News] SK hynix Reportedly Weighs Intel for HBM4E Base Dies amid TSMC Cost Pressure and Supply Diversification

(Photo credit: Intel)