Is Apple having the last laugh on AI?

What looked like falling behind now looks smart

By Jackie Snow, Quartz Media

Dec 16, 2025

Apple's AI efforts have been a mess. The company that once defined consumer technology spent 2025 playing catch-up, delaying promised Siri features and relying on Google's Gemini to fill gaps in its own Apple Intelligence platform. Wall Street punished the stock through the first half of the year as investors worried the iPhone maker had missed the defining technology shift of the decade.

Then the executive departures started. In one week this month, AI chief John Giannandrea retired after delays made his position untenable, the head of interface design left for Meta, and the company announced retirements for its general counsel and environmental chief. Dozens of engineers have bolted for competitors.

OpenAI acquired Jony Ive's design studio for more than $6 billion and has since hired dozens more Apple engineers. Meta used massive compensation packages to poach AI researchers, including the team that worked on Apple's AI models. All this happens as chief executive Tim Cook, who took over from Steve Jobs in 2011 and is now 65, faces growing succession planning discussions among analysts and investors.

And yet Apple's stock has surged about 40% since June, giving the company a $4.1 trillion market cap second only to Nvidia. Meta and Microsoft have watched their stocks slide despite pouring billions into AI infrastructure. Apple is now the second most expensive stock by forward P/E ratio in the Magnificent Seven, trailing only Tesla.

Sponsored

Apple is playing a different game

What looked like falling behind now looks smart: Apple skipped the AI arms race and the massive spending that comes with it. Instead, at WWDC earlier this year, the company announced the Foundation Models framework, which lets developers access Apple’s on-device AI with just three lines of code.

This could supercharge AI adoption by solving three problems. API costs make AI features prohibitively expensive for smaller developers. Requiring an internet connection limits where and when AI features work. And companies worry about liability and data security when sending user information to the cloud. Apple's on-device approach eliminates all three barriers, potentially unlocking widespread AI adoption that hasn't materialized despite billions in investment.

Apple's 3 billion parameter model can't compete with frontier systems like GPT-4, which is rumored to be over a trillion parameters. But that misses the point. With more than a billion iPhone users and millions of developers in its App Store ecosystem, Apple is betting that widespread access to decent AI beats having the most powerful model that few can afford to use. Even if Apple's current AI models are inferior to OpenAI's or Google's, enabling millions of developers to experiment with free, on-device AI could generate the breakthrough applications that Apple hasn't been able to create internally.

Apple is betting it can win AI the same way it won mobile. The company doesn't need the most powerful AI models if it has the best AI apps. Millions of developers experimenting with on-device AI could make the iPhone the best place to experience AI. And it's significantly cheaper than building AI models that might be obsolete by next quarter.

As one portfolio manager told Bloomberg, Apple has "kept their heads and are in control of spending, when all of their peers have gone the other direction." Investors increasingly question whether AI investments will pay off.

The chaos might be overstated

Some departures were overdue. Giannandrea's exit acknowledges what many knew about Apple Intelligence's struggles. Others could be upgrades, like the new head of interface design who has a background in the field replacing someone who came from fashion and advertising. And others seem to be just bad timing for the news cycle not signs of fleeing a sinking ship, like the general counsel and environmental chief retiring after long tenures.

The talent losses are real, particularly in AI research, where Meta and OpenAI are aggressively recruiting. But Apple confirmed this week that Johny Srouji, whose teams developed the M-series processors and in-house cellular modems, is staying, despite rumors he was considering leaving.

Apple still faces real challenges. The company hasn't launched a successful new product category in a decade. Its AI products have disappointed. Leadership transitions create uncertainty when it needs to prove it can compete in the AI era.

But Cook remains in control, the executive team is being refreshed with internal talent, and the company isn't overpromising on AI features it can't deliver. Whether Apple's measured approach proves correct depends on execution.

But for a company supposedly in crisis, the market is betting on stability.

—Jackie Snow, Contributing Editor