After years of watching OpenAI, Google, and Anthropic eat its lunch, Apple is finally doing what it should have done in 2017.
The Summary
- Apple is overhauling Siri as part of iOS 27, with a redesigned experience and standalone Siri app aimed at competing with ChatGPT
- Bloomberg's Mark Gurman says consumers "should be pumped" about the changes, calling them "massive" in Apple's effort to bring AI to the masses
- Details expected at WWDC on June 8, marking Apple's staged comeback in digital assistants after years of falling behind
The Signal
Apple spent a decade convincing people that talking to their phones was normal, then went silent while everyone else figured out what to do with large language models. Now, with iOS 27 arriving later this year, the company is trying to course-correct. The overhaul includes both a complete Siri redesign and, notably, a standalone Siri app, a signal that Apple is treating this as a platform play, not just a feature update.
TechCrunch reports new renders showing what this looks like in practice. A standalone app positions Siri as a direct ChatGPT competitor, something people open deliberately rather than summon by accident while trying to turn down their volume. That matters. The difference between an assistant you invoke and an assistant you visit is the difference between a tool and a workspace.
"Consumers should be pumped about the massive changes Apple is making to Siri."
The timing is revealing. Apple has 2 billion active devices and has been hemorrhaging mindshare to OpenAI for two years. Every iPhone user who's installed ChatGPT is a referendum on Siri's inadequacy. Apple is positioning this as bringing "artificial intelligence to the masses," but the masses already have AI. What they don't have is AI that works natively with their photo library, messages, calendar, and the rest of the Apple stack without jumping through API hoops.
That's the real play here. Not better answers to trivia questions, but an agent that actually lives in your device ecosystem. Apple has always won on integration, and they've had years to watch what people actually do with LLMs. The question is whether they learned the right lessons.
What makes this different from previous Siri updates:
- Standalone app suggests a hub model, not just voice commands
- Direct positioning against ChatGPT indicates conversational AI, not task automation
- "Massive changes" language from Gurman, who rarely oversells Apple rumors
The Implication
Watch what Apple announces at WWDC on June 8. If they ship a conversational interface that can actually orchestrate tasks across native apps without breaking, they'll have built what OpenAI can't: an agent with privileged access to the data that matters. If they ship another voice interface with party tricks, they'll confirm what many suspect, that they're a hardware company that got caught flat-footed on the agent economy.
The real test is whether this Siri can do the boring, repetitive stuff people actually want automated. Not "what's the weather," but "reschedule my 3pm, update the shared doc, and tell me what I missed while I was offline." Apple has the distribution. They've finally admitted they need to compete. Whether they can actually build an agent worth using is about to become very public.