Apple has made a big choice in the fast-moving AI race. For its next major Siri upgrade, Apple picked Google’s Gemini AI instead of ChatGPT-maker OpenAI. The decision sends a clear message about who Apple trusts most to power one of its most important products.
For OpenAI, this feels like a hard loss. Not just because of money, but because of what it says about where the company stands right now.
Apple’s big AI decision
On Monday, Apple confirmed that Google’s Gemini models will power its new AI features, including a smarter version of Siri. These updates are part of what Apple calls “Apple Intelligence” and are expected to roll out starting this spring.
Apple said it spent nearly a year testing different options. In the end, it decided Google’s AI was the most reliable and capable choice. The deal is set to last several years and will use Google’s technology and cloud systems, though Apple says user data will stay protected on its own devices and private servers.
Apple also made it clear that its existing ChatGPT feature is not going away. Users will still be able to ask ChatGPT certain questions through Siri. But that role now looks small. ChatGPT is no longer the main brain behind Siri’s biggest upgrade in years.
Why this hurts OpenAI
On paper, OpenAI still looks strong. ChatGPT has around 800 million weekly users. It helped start the current AI boom. But Apple’s choice shows that popularity alone is not enough.
When Apple needed an AI partner it could fully depend on, it chose Google.
That matters because Apple is extremely careful. This is the same company that canceled products after announcing them because they did not meet its standards. Apple delayed Siri’s AI upgrade publicly and admitted it was not ready. It does not make decisions like this lightly.
So when Apple picked Gemini over ChatGPT, it sent a quiet but powerful signal. OpenAI is no longer seen as the safest bet for large, core products where reliability matters most.
Apple tried to do it alone first
Apple did not rush to Google right away. At first, it wanted to build its own AI models. At WWDC 2024, Apple showed off a new Siri that could understand personal context, search messages and emails, and take actions across apps.
The demos looked impressive. But behind the scenes, things were not working well.
Apple later admitted the system was not reliable enough. The AI struggled to handle everyday tasks and complex language at the same time. Eventually, Apple delayed the feature and reshuffled its leadership. The team responsible for Siri was moved under top executives Craig Federighi and Mike Rockwell.
There were also signs of trouble inside Apple’s AI teams. Key leaders left for rivals like Meta, and others started looking elsewhere. The message was clear. Apple needed outside help.
A careful search for the right partner
Once Apple accepted that it needed a partner, it ran detailed tests. It looked at three main options: OpenAI, Anthropic, and Google.
Apple built two internal systems. One used its own AI models. The other used outside technology. Both were tested against Apple’s strict quality rules.
Anthropic’s Claude impressed Apple early on. Engineers liked its safety and performance. But pricing became a problem, and talks slowed down.
Google, meanwhile, began training a custom version of Gemini that could run on Apple’s servers. Google already had experience powering AI features on Samsung phones. It also had the scale, money, and infrastructure Apple wanted.
OpenAI was part of the process the whole time. In fact, it already had a head start. ChatGPT was integrated into iOS 26, and Apple even ran ads promoting it. But that early advantage did not last.
Also Read: Judge Clears the Way for Jury Trial in Musk vs Altman OpenAI Lawsuit
OpenAI’s bad timing
The timing could not have been worse for OpenAI.
In August, OpenAI released GPT-5. Many users were disappointed. They said it felt colder, slower, and struggled with simple questions. Instead of building excitement, the release raised concerns.
By December, OpenAI CEO Sam Altman sent a “code red” message to staff. He admitted ChatGPT had serious issues with speed, reliability, and personalization. Other projects were paused so teams could focus on fixing the core product.
Daily emergency meetings started. That is not the position you want to be in while competing for a massive deal with Apple.
At the same time, Google was moving in the opposite direction. In November, it released Gemini 3, which topped many industry benchmarks. Google’s cloud business was winning large deals, and its stock surged through 2025.
The contrast was sharp. One company was fixing problems. The other was building momentum.
Money and stability matter
There is also a money gap that is hard to ignore.
OpenAI burns huge amounts of cash every year and is still not profitable. By its own estimates, it needs massive future revenue just to break even. It relies heavily on investors.
Google, on the other hand, funds its AI work through advertising and cloud profits. It can afford to spend for the long term. That stability matters when AI requires constant investment in chips, data centers, and top talent.
For Apple, choosing Google reduced risk. For OpenAI, it highlighted a weakness.
More than just lost revenue
Reports suggest Apple may pay Google around $1 billion per year for Gemini access. That is a lot of money, but the real loss for OpenAI goes deeper.
This was a chance to become the core AI engine for hundreds of millions of iPhone users. It would have meant steady income, scale, and trust from one of the world’s biggest tech companies.
Apple could have chosen OpenAI to avoid working more closely with Google, a direct rival in smartphones. It chose Google anyway.
That decision makes OpenAI’s road ahead harder. Enterprise customers are important, but consumer partnerships like Apple’s offer long-term stability. Losing that opportunity raises questions about OpenAI’s future growth.
A shift in the AI race
There is some irony here. Last year, Apple hinted it might let users choose between different AI models in the future. Now, Google has taken the center spot, while OpenAI watches from the side.
Two years ago, ChatGPT dominated headlines and defined the AI conversation. Today, the race looks different. Big tech companies with deep pockets are pulling ahead.
Apple’s choice does not mean OpenAI is finished. But it does show that the AI race is not just about being first. It is about being reliable, stable, and ready when it matters most.
