Apple plans to launch its long-promised overhaul of Siri in September, and it’s leaning on two outside heavyweights to get there: Google and Nvidia. That’s according to The Information, which reports the revamped assistant will arrive alongside Apple’s usual fall product cycle. What stands out here is the admission baked into the news. Apple, a company that prefers to build its own everything, is bringing in help.
What’s actually happening
The new Siri is the assistant Apple first teased back in 2024 as part of “Apple Intelligence.” It was supposed to understand context, act across your apps, and handle messy real-world requests. It didn’t ship on time. The feature slipped repeatedly, leadership got reshuffled, and Siri became the symbol of Apple playing catch-up in AI.
Now, per The Information, the September launch will lean on:
- Google, whose Gemini models help power the smarter language understanding behind the new Siri.
- Nvidia, whose chips support the heavy compute needed to train and run those models.
In plain terms: Apple is renting brains and muscle it couldn’t build fast enough on its own.
Why this matters
This is significant because it breaks Apple’s usual script. For years Apple sold the story that owning the whole stack, silicon, software, and services, was its edge. Handing the core intelligence of Siri to Google’s models is a real concession. It says Apple decided shipping a competent assistant now beats waiting years to do it all in-house.
It also reshuffles the competitive board. Google already pays Apple billions to be the default search engine in Safari. If Gemini also becomes the engine behind Siri, Google moves even deeper into the iPhone, the most valuable device platform on earth. That’s a strong position for Google and an uncomfortable dependency for Apple.
For Nvidia, it’s more confirmation that the AI buildout runs through its chips. Even Apple, which designs its own silicon, reportedly needs Nvidia hardware to train at the scale modern assistants demand.
How we got here
The status quo before this was rough for Apple. The original AI Siri was announced, demoed, delayed, and quietly pushed into 2026 territory. Rivals didn’t wait. Google shipped Gemini across Android and its apps. OpenAI’s ChatGPT became the default voice people reach for. Amazon rebuilt Alexa with generative AI. Siri, meanwhile, still struggled with basic follow-up questions.
Bringing in partners is how Apple closes that gap quickly. It’s not elegant by Apple standards, but it’s pragmatic.
What to watch
A few things will tell you whether this works:
- How much is Apple, how much is Google. Apple will likely frame Siri as its own product with privacy guarantees, even if Gemini does heavy lifting underneath. Watch how clearly it discloses that.
- Privacy handling. Apple’s brand rests on keeping data on-device. Routing requests to outside models raises obvious questions. Expect Apple to push private cloud processing hard in its messaging.
- The regulatory angle. The existing Google-Apple search deal is already under antitrust scrutiny. A second deep Google integration could draw more attention.
- Whether it actually ships in September. This feature has slipped before. A firm date is progress, not a guarantee.
The takeaway for practitioners
If you build on Apple’s platforms, plan for a more capable assistant layer arriving this fall, with app-intent style actions becoming more realistic to target. If you’re tracking the AI market, this is another data point that the “build it all yourself” era is giving way to partnerships, even among the companies most committed to going it alone.
Apple admitting it needs help is the real story here. The assistant is the product. The partnership is the signal. More details are available in the original report from The Information.