Intel’s Ohio Fab Is Quietly Hitting Every Major Milestone – And 2028 Could Be Explosive
Monday, December 01, 2025Intel’s Ohio Fab Is Quietly Hitting Every Major Milestone – And 2028 Could Be Explosive
For years, Intel’s massive Columbus, Ohio “mega-fab” complex has been written off by skeptics as a perpetually delayed boondoggle. Headlines screamed “2030 at the earliest” or even “never” after repeated slips in the original schedule.
But a detailed thread posted on X this weekend by analyst Lukateake (who has a very strong track record on Intel Foundry leaks) paints a dramatically different picture: the Ohio site is not only on track — it’s quietly crushing every major technical and funding milestone needed for high-volume 14A (1.4 nm-class) production in late 2028.
The Milestones Nobody Is Talking About
- High-NA EUV tools already installed and qualified
Intel’s first production High-NA scanner is running real 30,000-wafer quarters in Ohio right now — not lab toys, actual volume-capable throughput. - $8 billion CHIPS Act grant + 10% U.S. government equity stake
Money is signed and flowing; Washington now has literal skin in the game. - Fab shells and infrastructure basically done ahead of schedule
The first two fabs are structurally complete; the rest of the six-fab campus is well underway.
The 2028 Ramp Looks Very Real
Intel is internally guiding Q4 2028 for first 14A high-volume manufacturing in Ohio at 2,000–7,000 wafers per month to start, scaling aggressively in 2029.
Reported or late-stage customers include:
- Apple – entry-level M5/M6 chips
- AWS – next-gen Graviton & Trainium AI chips
- NVIDIA – 14A risk starts with possible volume shift (Blackwell-ultra / Rubin)
- Microsoft & Google Cloud custom silicon (already 18A customers)
Why the Old Narrative Is Still Stuck in 2022
Most of the doom articles are recycling talking points from when Intel was years late on 10 nm and the original 2025 Ohio plan looked impossible. Since then Intel has shipped Intel 4, Intel 3, and 18A on time or early, restructured the foundry, and locked in real external customers + CHIPS money.
Bottom Line
The buildings are up. The High-NA tools are running wafers. The money is in the bank. The customer pipeline is the strongest it has ever been.
If Intel executes — still an “if,” but a much smaller one than 24 months ago — the United States will have its first true leading-edge fab outside Asia by 2028–2029.
And the first company shipping commercial 1.4 nm chips in volume might just be wearing blue.
2028 is going to be a lot more interesting than the bears think.