Behind closed doors at Apple, Siri is getting a rethink that could change how we talk to our devices. Apple's been quietly working on a ChatGPT-like app to help its engineers test the overhauled version of Siri. The app isn't going to be released to the public, and it's strictly for internal testing, but it offers a rare peek at where Siri is headed.
The testbed signals something bigger. Apple knows the current Siri architecture is out of its depth in a conversational AI world. The app is being used to test new Siri features such as more contextual awareness, an ability to do more in and between apps, and deeper integration with personal data. Think picking up a grocery list in Notes, booking a delivery window in a shopping app, then dropping reminders on your calendar without losing the thread. Not just better small talk, smarter workflows.
What's really happening behind the scenes
Here's the timeline. Apple has been working on a smarter version of Siri since the launch of iOS 18. The company initially planned to debut an Apple Intelligence Siri as part of the update, but delayed the functionality until 2026. The early architecture buckled under context and conversation demands, so Apple hit pause and rethought the foundation.
Apple couldn't get the first-generation Siri architecture that it was using to work properly, so Siri needed to be rebuilt from the ground up using a second-generation architecture that relies on large language models. That meant redoing conversation management, long-term context, and the framework that ties into apps. The Apple Intelligence Siri plan was scrapped, and Apple decided to overhaul Siri with second-generation architecture, speeding up an LLM transition.
The internal testing app is the lab bench. The app essentially takes the still-in-progress technology from the new Siri and puts it in a form employees can test out more efficiently. It also helps Apple gauge whether a chatbot-style interface has legs, allowing the company to gather feedback on whether the chatbot format has value. Faster iteration, clearer signals, fewer guesswork detours.
The LLM transformation coming to your pocket
Bottom line, the next Siri will not feel like the Siri you know. The upcoming version of Siri will use advanced large language models like ChatGPT, Claude, Gemini, and other AI chatbots. The result, LLM Siri will be able to hold continuous conversations, provide human-like responses to questions, and complete more complex tasks.
The testing app points to how that plays out. It lets Apple users manage multiple conversations across different topics and can save and reference past chats, follow up on earlier queries and support extended back-and-forth exchanges. Plan a vacation, then come back days later to swap flights and tweak the hotel. Start a research thread on iPhone, pick it up on Mac, then reference the whole thing when you add a reminder. No more groundhog day with every query.
That memory gets sharper when combined with Apple's "World Knowledge Answers" system. Apple is not just cloning a chatbot. They are building World Knowledge Answers internally, with some Apple executives referring to it as an "answer engine". It will include an interface that supports text, photos, videos, and local points of interest, with Siri able to summarize search results to provide a clear summary of content.
That positions Siri against Google's AI overviews and tools like Perplexity. Not just timers and lights anymore. A hub that can understand layered questions across formats and give you a usable answer.
The partnership puzzle: who's powering the future?
Strategically, Apple is playing it wide. Apple has held talks with Anthropic, OpenAI, and Google, and may use a third-party company to power the new version of Siri rather than relying solely on its in-house AI models.
The eye-opener, Apple and Google have signed a formal agreement that will see Apple evaluating and testing a custom Google-designed Gemini AI model that could power some of the Siri features. At the same time, Apple is still testing models designed by Anthropic and its own in-house models for the planner functionality, and they could use Google's models there too.
This multi-vendor mix lets Apple tune strengths, Google's search knowledge, Anthropic's reasoning, Apple's privacy-focused data work, without giving up control of the experience. Inside that setup, Apple's own Foundation Models will be used for searching user data, so sensitive information stays inside the house.
Pragmatic, not precious. Build what matters in-house, borrow what accelerates the roadmap, keep the user experience and privacy intact.
When can you actually use this?
Here is the roadmap. Apple is on track to launch the LLM version of Siri in early 2026, likely as part of an iOS 26.4 update that could come in March. That lands about a year after the original target, a full year after Apple originally wanted to launch the updated version of the personal assistant.
More waves are queued up. Apple plans to unveil a new look for Siri at the end of next year, giving it a more humanoid design similar to the Mac's finder logo. Then, later in 2026, Siri will get a visual redesign and a built-in health feature that will be the backbone of a paid wellness subscription service.
Search will show up first in familiar places. You will see it in Siri, Safari, and Spotlight, turning them into touchpoints for Apple's answer engine.
Yes, that cadence puts Apple behind faster-moving rivals. It also gives them time to make the experience feel native, not bolted on.
What this means for the Apple ecosystem
This is not just "Siri, but smarter." It is Apple trying to make conversation the glue across tasks. The current ChatGPT hook inside Apple Intelligence feels like a guest appearance. LLM Siri is Apple taking the mic.
There is a twist. Apple executives have expressed doubt in the usefulness of chatbot interfaces for artificial intelligence features. Craig Federighi saying Apple wants to build an experience that is "integrated into everything you do, not a bolt-on chatbot on the side" at WWDC. Which makes the internal testing app a telling move. Apple is checking whether a classic chat window earns its keep, even as they design for integration first.
If it works, productivity changes shape. You keep a single conversation that runs through research, planning, execution, and follow-up. Less app-juggling, more momentum. That is the kind of polish people buy into the ecosystem for.
Where do we go from here?
The internal ChatGPT-like app shows Apple is taking the assistant race seriously. Apple has reportedly scrapped plans to launch a separate chatbot app, choosing instead to embed AI into its native search layers, yet they are not closing the door on conversational interfaces.
The 2026 window feels far away when others ship monthly. Still, the method is familiar, test quietly, shape the edges, launch when it feels like Apple. Can they keep privacy and tight integration while catching rivals on capability? That is the ballgame.
If you live on Apple devices, this could be the biggest shift since the iPhone era. A Siri that remembers, understands context, and moves between apps with you. The assistant Apple hinted at over a decade ago, finally within reach.
Comments
Be the first, drop a comment!