The news of Apple’s pending deal with Google says a lot. While neither Apple nor Google will likely ever publicly acknowledge it beyond necessary financial disclosures, Apple fans should take note. It’s both very good news for users and also somewhat concerning. We should be equally pleased and worried.
### The Deal
According to reports (Apple has not officially confirmed), Apple will pay Google about $1 billion a year to use a customized version of Google’s Gemini AI model for the new Siri, expected to be released in the spring. This model is large and advanced, reportedly boasting 1.2 trillion parameters, and will run on Apple’s own Private Cloud Compute servers, ensuring neither Google nor any third party can access your data.
The partnership reportedly came after Apple evaluated Google’s AI alongside offerings from Anthropic, OpenAI, and its own internally developed large language model (LLM) technology.
### What It Means for Users
For users, this is mostly good news. The latest versions of Gemini rank among the top LLMs in the industry—benchmarks vary, and Apple’s version might differ from Gemini 2.5 Pro, but it’s clear Apple is not settling for a second-rate model.
Of course, talking to the new Siri won’t be the same as interacting with Google’s Gemini directly. The voices will sound different, and Siri will have distinct priorities and tuning. Plus, Siri will have access to your device’s private data to provide a more personalized experience. You can think of it as two different cars powered by the same engine but with different designs and features.
Apple’s willingness to spend a significant sum to integrate another company’s core technology into one of its most important—and often criticized—features signals a notable shift in Cupertino’s mindset. Traditionally, Apple builds or acquires technology when it needs something new, often both. That Apple is breaking from its “homegrown only” approach to deliver a better Siri is commendable.
### Not Invented Here
However, Apple fans should also approach this with caution. Apple’s “Not Invented Here” ethos is well-known—where the company feels the need to own or build everything, sometimes regardless of whether it benefits users.
There are examples where it has paid off. For instance, after years of effort, Apple’s cellular modems and N1 networking chips now provide experiences on par with Qualcomm and Broadcom. But there are also cases where Apple’s stubborn in-house focus wasn’t beneficial.
When the OpenGL graphics API became outdated, Apple could have adopted the open Vulkan standard, helping influence its future development. Instead, Apple created its own Metal API, which hasn’t necessarily been better for developers or users.
Similarly, Apple didn’t need its own lossless audio format. And perhaps most notably, Google pays Apple about $20 billion annually to remain the default search engine in Safari. While other search engines are available, most users stick with the default, making this lucrative for Google—but not necessarily good for users.
Google’s search experience has steadily declined over time, leveraging data from these searches to consolidate control over search and web advertising. Apple should have made a serious investment years ago to build its own privacy-minded, ad-free web search.
We’ve also seen the pitfalls of Apple handling core services in-house with Apple Maps. Initially, Apple’s attempt to move away from third-party mapping resulted in a poor product, largely because it lacked ownership and control over its data sources. It took years for Apple to develop a high-quality Maps experience based entirely on its own data, and now it is top-tier.
### Apple Needs Its Own LLM
It’s clear that some core technologies Apple must build and fully control, while others can be sourced externally.
A foundational AI large language model is unquestionably in the former category. Over the coming years, AI models will be integrated into far more than chatbots. Apple’s products already use AI extensively—from cameras that recognize gestures to image editing and notification summaries.
However, the foundational LLM that interacts with users, manages devices, and gathers information is the most critical AI model in the ecosystem. Apple having its own top-tier LLM is as important as controlling any other major technology component—and arguably more important than owning its own web browser.
It’s positive that Apple recognizes its internal LLM technology isn’t yet competitive and is willing to collaborate with Google to improve Siri. But ultimately, Apple must catch up to or surpass the technology it is buying.
That may be a challenge, considering competitors are rapidly advancing their LLMs, while Apple seems to be losing AI talent consistently.
### Conclusion
As a user, you should be glad about the Apple-Google Gemini-Siri AI deal—for now. It promises a big leap forward for Siri and user experience. But it’s critical for Apple to continue investing in its own AI capabilities. Otherwise, this partnership may only be a temporary fix rather than a long-term solution.
https://www.macworld.com/article/2964789/apples-1b-google-ai-deal-will-be-great-for-iphone-users-until-its-not.html