By Zaheer Kachwala and Anhata Rooprai
June 24 (Reuters) - Apple turning to Intel for chips, as Washington announced last week, has the neat logic of necessity meeting ambition. But it is not that simple, as analysts say any advanced Intel chip will take two to three years to make and even longer to translate into gains due to the long and exacting production process.
A deal - which neither company has formally announced - would pair Intel's effort to rebuild its credibility as a contract chipmaker
with Apple's search for more manufacturing capacity, as its supplier TSMC struggles to meet surging AI chip demand from the likes of Nvidia.
Supply constraints at the contract manufacturer have held back iPhone sales, Apple CEO Tim Cook said in April.
Baked into this deal is a strategic calculation. Intel has emerged as a key pillar in the U.S. plan to rebuild domestic chipmaking through tariffs and incentives, thanks to its 10% stake in the company and a $5 billion investment from Nvidia at the behest of President Donald Trump.
"The absolute best possible case would be 2-3 years before the first chips flowed off the line. It takes 2 years to design an SoC (system on chip) of this complexity, and a further 4 months through production cycle time to volume ramp up," said Malcolm Penn, CEO of chip research firm Future Horizons.
This assessment assumes Intel's technology is fully worked out and its design tools are reliable enough for Apple to depend on, Penn said. "With no track record, that's a huge leap of faith and commercial and financial risk," said Penn, who termed the deal "a shotgun wedding".
FIRST TESLA, NOW APPLE
After missing out on the early stages of the AI boom, Intel has begun to show tentative progress, landing Tesla as a customer in April and positioning itself for a more consequential partnership with Apple.
Analysts are divided on which Intel manufacturing process Apple will choose.
Some see it following Tesla onto Intel's next-generation 14A, a process years away from volume production but built on the world's most advanced chipmaking tools.
Others expect Apple to sacrifice cutting-edge gains for reliability, favoring 18A-P, a refined version of Intel's most advanced process that began initial production this month - or an older, reliable node like Intel 3.
"Apple would probably want to use Intel's 14A process technology... and that's expected to be available in 2028 or 2029 so it's still going to be a while," said Bob O'Donnell, an analyst at TECHnalysis Research.
"However, if it proves to be true, it's an extremely important development for Intel's foundry business and US-based semiconductor manufacturing in general."
Daniel Newman, CEO of tech research firm Futurum Group, said volume production of Apple-designed chips was unlikely until late 2027 or early 2028, with the initial work focused on less critical components used in MacBook Air or some iPad Pro models.
INTEL HAS FACED POOR CHIP YIELDS
Apple may even hedge, testing Intel with lower-end products before committing its most critical chips, analysts said.
Intel, which has historically faced issues with the timeline and quality of its chips, will have to meet Apple's high expectations for yield, a standard that the world's largest consumer electronics company has come to expect from TSMC. Yield is the percentage of chips on a silicon wafer that work correctly when manufacturing is done.
"Investors are pricing in perfect execution by Intel, which is a company that hasn't delivered for about 20 years. Granted, it looks like Intel has made strides with its latest manufacturing process, but I think we should all at least modestly discount a perfect outcome," said Paul Meeks, head of tech research at Freedom Capital Markets and an Intel investor.
(Reporting by Zaheer Kachwala and Anhata Rooprai in Bengaluru; Editing by Sayantani Ghosh and Arun Koyyur)
