Abner Li writing for 9to5Google: Apple Might Use Google Gemini to Power Some AI Features on the iPhone

According to Bloomberg, there are “active negotiations to let Apple license Gemini, Google’s set of generative AI models, to power some new features coming to the iPhone software this year.” Apple has also talked with OpenAI, which powers Microsoft’s AI capabilities.

I’m seeing people on social media going nuts over this rumor, with people complaining about things such as working with Google being antithetical to Apple’s values to Apple being so inept at generative AI that they’ve given up and are outsourcing it all to Google.

Take a breath…it’s a rumor, and good lord are people extrapolating whatever storyline they want to be true here.

First, Apple and Google have a long established relationship working together on iPhone features. Google pays Apple well north of $10 billion per year to be the default search engine on Apple devices, as well as being used in things like Spotlight and Siri to provide web results when looking things up. Apple may throw shade at Google, but they certainly aren’t above working together, and in my opinion that’s not a bad thing overall; Google is still the best search engine for most people and it makes Apple’s features better.

Second, there is absolutely no evidence that Apple has somehow “failed” at building what they want for running LLMs and other ML features on their devices. The company has published numerous papers over the years, including a big one literally the same day that the Gemini rumor hit.

All rumors should be taken with a grain of salt, and even if the rumor is true, we don’t know how it fits into the overall picture of what Apple’s generative AI storyline will be. Maybe they have completely failed to develop the features they want and are white-labeling Google’s Gemini, but there’s no use in getting all up in arms until at least WWDC in June.