Apple AI in 2026: Siri 2.0 Is Still “Coming Soon” (and That’s a Problem)
Apple has a Siri problem. While Google, OpenAI, and Anthropic have been shipping AI features at breakneck speed, Apple keeps promising a revolutionary Siri upgrade that keeps getting delayed.
The latest: Siri 2.0 was supposed to arrive in spring 2026. Then it got pushed to “later this year.” Bloomberg sources say it might not ship until September — or later.
For a company that prides itself on polish and timing, this is embarrassing.
What Apple Promised
Back in June 2024, Apple previewed Apple Intelligence and a dramatically improved Siri with three major new capabilities:
Onscreen awareness: Siri would understand what’s on your screen and act on it. “Send this article to Mom” while reading a news story. “Add this event to my calendar” while looking at an email.
Personal context: Siri would understand your personal information — your relationships, preferences, schedule — and use it to provide more relevant responses.
Cross-app actions: Siri would be able to complete multi-step tasks across multiple apps. “Find the photos from last weekend’s hike and create a shared album for the hiking group” — one command, multiple apps, done.
These features would have made Siri competitive with Google Assistant and ChatGPT. The problem is they’re still not here.
Why the Delays?
Apple hasn’t officially explained the delays, but the reasons are pretty clear:
Quality bar. Apple’s AI features need to work reliably on-device, not just in the cloud. Running sophisticated AI models on an iPhone requires optimization that cloud-first companies don’t have to worry about.
Privacy constraints. Apple’s commitment to on-device processing means they can’t just throw everything at cloud servers like Google does. This limits what Siri can do and makes development harder.
Integration complexity. Making Siri work across every Apple app and third-party apps is an enormous engineering challenge. Each app has different APIs, data formats, and interaction patterns.
The LLM gap. Apple’s in-house language models aren’t as capable as GPT-4 or Claude. They’ve partnered with OpenAI for some features, but relying on a competitor for core functionality is awkward.
What Apple Intelligence Actually Does Today
While Siri 2.0 is delayed, Apple Intelligence has shipped some features:
Writing tools: Rewrite, proofread, and summarize text across apps. These work well and are genuinely useful.
Image generation: Image Playground and Genmoji let you create images and custom emoji. Fun, but not transformative.
Notification summaries: AI-generated summaries of notification stacks. Useful when you have dozens of notifications, but the summaries are sometimes hilariously wrong.
Photo features: Clean Up (remove objects from photos), natural language photo search, and Memory Movies. The photo features are probably the best Apple Intelligence features so far.
Mail summaries: AI-generated email summaries and smart replies. Similar to what Gmail has had for a while.
The honest assessment: these features are nice but incremental. None of them are the “wow, this changes everything” moment that Apple needs to compete with Google and OpenAI.
The Competitive Gap Is Growing
Here’s what worries me about Apple’s AI strategy: the gap between Apple and its competitors is growing, not shrinking.
Google has Gemini integrated into Search, Workspace, Android, and Chrome. OpenAI has ChatGPT with 200+ million users. Anthropic has Claude powering enterprise workflows. Microsoft has Copilot everywhere.
Apple has… writing tools and notification summaries.
The longer Siri 2.0 takes, the more users get accustomed to using ChatGPT, Google Gemini, or Claude for the tasks that Siri should be handling. Once habits form, they’re hard to break.
The Privacy Advantage (If They Can Ship)
Apple does have one genuine advantage: privacy.
If Apple can deliver a Siri that’s as capable as Google Assistant but processes everything on-device, that’s a compelling value proposition. Many users would prefer an AI assistant that doesn’t send their data to cloud servers.
But “if they can ship” is doing a lot of heavy lifting in that sentence.
What to Expect for the Rest of 2026
My predictions:
Siri 2.0 ships in September 2026 with iOS 19. Apple will tie it to the iPhone 17 launch for maximum marketing impact. It’ll be good but not as revolutionary as the 2024 preview suggested.
Apple Intelligence features will expand gradually. More writing tools, better photo features, and improved notification handling. Incremental improvements, not major changes.
The OpenAI partnership deepens. Apple will continue relying on OpenAI for the most capable AI features while building their own models for on-device tasks.
Apple won’t catch up in 2026. The AI gap between Apple and Google/OpenAI will narrow but not close. Apple’s strength has never been being first — it’s been being best. The question is whether “best” still matters when competitors are years ahead.
Should You Wait for Apple?
If you’re an Apple user wondering whether to start using ChatGPT or Google Gemini while waiting for Siri to improve: don’t wait. Use the best tools available now.
Apple will eventually ship a competitive AI assistant. But “eventually” could be late 2026 or even 2027 for the full feature set. In the meantime, there’s no reason to limit yourself to Apple’s current AI capabilities.
The best AI assistant is the one that’s available and works. Right now, that’s not Siri.
🕒 Last updated: · Originally published: March 12, 2026