Product Launch

Siri AI Finally Arrives in iOS 27 Beta, But the AI World Has Moved On

Apple's long-delayed Siri AI has finally arrived in the iOS 27 beta, offering natural conversations, personal context, and photo/document retrieval. The launch feels anticlimactic as the AI industry has moved ahead, but the improvements are significant, powered by a partnership with Google's Gemini models. The official release is expected in September.

Neura News

Neura News

Neura Market Editorial

August 3, 20268 min read
Siri AI Finally Arrives in iOS 27 Beta, But the AI World Has Moved On

Apple's long-delayed Siri AI has landed in the consumer beta of iOS 27, which launched in July. The assistant now understands personal context, taps into broad world knowledge, and surfaces relevant information stored on the iPhone. It feels functional and fairly impressive. Yet the launch feels anticlimactic because the AI race has surged ahead while Apple dawdled for months and years.

The new Siri allows natural back-and-forth conversations with adjustable expressivity and pacing of voice responses. For the first time, users can type to Siri or use it through a dedicated app, referencing past conversations. The assistant can find photos, emails, contacts, texts, or calendar appointments even when the user is unsure of specifics. For example, Siri can pull up the last receipt saved on the iPhone without the user knowing where it was stored or what it was for.

What the New Siri Can Actually Do

Siri can retrieve information from photos, such as a Driver's License number from a picture or a QR code from a screenshot. It can launch websites and use apps for directions, music, photo editing, email drafting, audiobooks, and more. The assistant consistently and accurately plays the requested song, podcast, or audiobook on the preferred app. That is a low bar, but one where Siri previously stumbled.

The assistant can also brainstorm, refine ideas, learn about new topics, and explore recipes, such as suggesting something from random fridge ingredients. It answers questions about events or band performances based on location. Users can point the Camera's viewfinder at real-world objects to learn about them or split a restaurant bill. Siri AI is exactly what users always hoped Siri would be.

The only marvel is that Siri actually works after years of failing. Apple had multiple rounds of delays for Siri AI. The company has dawdled on AI for months and years. During that time, the AI industry advanced significantly. AI tools today are coding, building software, completing multi-step tasks, working alongside users, reasoning, thinking, remembering, and creating media. Being a functional chatbot or AI helper is no longer a breakthrough.

Apple's Secret Weapon: Google's Gemini

These improvements were made possible by Apple's partnership with Google for use of Gemini AI models. Apple used Google's technology to train and refine its own proprietary Apple Foundation Models. Apple did not just slap its name on Gemini AI. The company used Google's technology to train its own models. Those Apple Foundation Models run on Apple's own Silicon and Private Cloud Compute infrastructure.

The general public gains access to the new Siri when iOS 27 officially releases, expected in September. That timeline follows a beta that arrived in July. The delay between beta and official release gives Apple time to polish the experience. But the industry has already moved past simple chatbots. Siri AI delivered on Apple's promises and then some. Yet the launch feels more like fixing a longstanding bug than a revolutionary moment.

Sarah Perez, Consumer News Editor at TechCrunch, wrote the article covering the launch. She has been a TechCrunch reporter since August 2011. Before that, she spent over three years at ReadWriteWeb. Her background includes work in I.T. across banking, retail, and software. She can be reached at sarahp@techcrunch.com or via Signal at sarahperez.01.

Why the Launch Feels Anticlimactic

The AI industry has surged ahead during Apple's delay. Tools now handle complex, multi-step tasks and work alongside users in real time. They reason, think, remember, and create media. A functional chatbot is no longer a breakthrough. Siri AI is valuable and useful despite the anticlimactic feel. The improvements are significant and deliver on Apple's promises. The partnership with Google was crucial for training Apple's own models.

The launch feels more like a bug fix than a revolutionary moment. Siri AI's functionality is what Siri was meant to be all along. The assistant finally understands personal context and world knowledge. It can hold natural conversations and retrieve information from photos and documents. It can launch apps and play media correctly. These are the features users have wanted for years.

The only marvel is that Siri actually works after years of failing. Apple had multiple rounds of delays for Siri AI. The company has dawdled on AI for months and years. During that time, the AI industry advanced significantly. AI tools today are coding, building software, completing multi-step tasks, working alongside users, reasoning, thinking, remembering, and creating media. Being a functional chatbot or AI helper is no longer a breakthrough.

The #1 Newsletter in AI

Stay ahead of the AI curve

The most important updates, news, and content — delivered weekly.

No spam. Unsubscribe anytime.

What Comes Next for Siri AI

The official release of iOS 27 is expected in September. That will bring the new Siri to the general public. The beta has already shown what the assistant can do. Users can type to Siri or use it through a dedicated app. They can reference past conversations. The assistant can find photos, emails, contacts, texts, or calendar appointments. It can pull up receipts, retrieve Driver's License numbers, and read QR codes from screenshots.

Siri can launch websites and use apps for directions, music, photo editing, email drafting, audiobooks, and more. It can brainstorm, refine ideas, learn about new topics, and explore recipes. It answers questions about events or band performances based on location. Users can point the Camera's viewfinder at real-world objects to learn about them or split a restaurant bill.

The assistant consistently and accurately plays the requested song, podcast, or audiobook on the preferred app. That is a low bar, but one where Siri previously stumbled. Siri AI is exactly what users always hoped Siri would be. The only marvel is that Siri actually works after years of failing.

Apple's partnership with Google for Gemini AI models was key. Apple used Google's technology to train and refine its own proprietary Apple Foundation Models. Those models run on Apple's own Silicon and Private Cloud Compute infrastructure. Apple did not just slap its name on Gemini AI. The company used Google's technology to train its own models.

A Useful Assistant, Not a Revolution

Siri AI is valuable and useful despite the anticlimactic feel. The improvements are significant and deliver on Apple's promises. The partnership with Google was crucial for training Apple's own models. The launch feels more like a bug fix than a revolutionary moment. Siri AI's functionality is what Siri was meant to be all along.

The AI industry has moved beyond simple chatbots. Tools now code, build software, complete multi-step tasks, and work alongside users. They reason, think, remember, and create media. Being a functional chatbot or AI helper is no longer a breakthrough. Siri AI arrived after the AI industry moved beyond simple chatbots.

Sarah Perez has been a TechCrunch reporter since August 2011. She previously worked at ReadWriteWeb for over three years. Her background includes I.T. work in banking, retail, and software. She can be contacted at sarahp@techcrunch.com or via Signal at sarahperez.01.

TechCrunch Disrupt takes place October 13-15 in San Francisco. Attendees can save up to $330 on registration. The event will feature discussions on the latest in technology and AI. The article also includes an affiliate links disclosure, noting TechCrunch may earn a commission.

The general public gains access to the new Siri when iOS 27 officially releases, expected in September. That follows the beta launch in July. The new Siri understands personal context, taps into broad world knowledge, and surfaces relevant information stored on the iPhone. It allows natural back-and-forth conversations with adjustable expressivity and pacing of voice responses. Users can type to Siri or use it via a dedicated app for the first time, referencing past conversations.

Siri can find photos, emails, contacts, texts, or calendar appointments even when the user is unsure of specifics. It can pull up the last receipt saved on the iPhone without knowing where it was stored or what it was for. It can retrieve a Driver's License number from a pic or a QR code from a screenshot. It can launch websites and use apps for directions, music, photo editing, email drafting, audiobooks, and more.

The assistant consistently and accurately plays the requested song, podcast, or audiobook on the preferred app. That is a low bar, but one where Siri previously stumbled. Siri AI is exactly what users always hoped Siri would be. The only marvel is that Siri actually works after years of failing.

Related on Neura Market

More from Neura News

AI Models

42 Mathematicians Urge Royal Society to Warn Government and Media About AI Existential Risk

Forty-two mathematical fellows, including Fields Medal winners Martin Hairer, Peter Scholze, and Wendelin Werner, have signed an open letter urging the Royal Society to warn the UK government and media about existential risks from advanced AI. The letter follows recent breakthroughs in which leading models solved open research problems, including a Millennium Problem. None of the signatories are affiliated with AI companies. The group warns that AI labs' estimates of existential risk above ten percent must not be dismissed as hype, and that by the time the situation becomes obvious to the public, it may be too late to act.

Sep 18·2 min read
Developer

Steve Yegge Shuts Down Gas Town After Failing to Build Anything Else With It

Steve Yegge shut down Gas Town, his ultra-vibed coding agent orchestrator, after admitting he never built anything else with it despite heavy subscription spend. Databricks reported a 60% coding spend increase after rolling out GPT-6 Astra to 3,500 engineers, OpenAI published a misalignment disclosure framework with six case reports, and Xiaomi ran MiMo-V2.6 RL training in public with live telemetry.

Sep 18·21 min read