MODELS27 OCT 2026 · 18:13 UTC · 7 MIN READ

Apple Adds AI Assistant to Pages Keynote Numbers

Apple is rolling out a new AI assistant across Pages, Keynote, and Numbers that can design slides, draft text, restructure documents, calculate results, and visualize data by analyzing on-screen content and applying changes directly. The feature launches in beta for devices running iOS 27, iPadOS 27, or macOS 27 and later, using models from Apple and OpenAI. It also adds Model Context Protocol support for third-party AI agents.

MODELS DESK

Short answer: Apple is rolling out a new AI assistant across Pages, Keynote, and Numbers that can design slides, draft text, restructure documents, calculate results, and visualize data by analyzing on-screen content and applying changes directly. The feature launches in beta for devices running iOS 27, iPadOS 27, or macOS 27 and later, using models from Apple and OpenAI. It also adds Model Context Protocol support for third-party AI agents.

Apple AI assistant iWork Pages Keynote Numbers

Apple is rolling out a new AI assistant across its core productivity suite, bringing generative capabilities to Pages, Keynote, and Numbers. The feature arrives as part of a broader push that also touches creative applications like Pixelmator Pro, Logic Pro, and Final Cut Pro. According to the company, the assistant draws on models developed both internally and by OpenAI, and it is designed to understand the context of a document so it can take action rather than simply answer questions.

In Keynote, the assistant can design slides, add animations, and refine existing decks based on a user's intent. Pages gains the ability to draft fresh text, restructure sections, and tweak wording across a document. Numbers users can ask the system to calculate results, visualize trends, organize tables, and even recommend chart types that fit the data at hand. The company says these functions work by analyzing the content already on screen, then applying changes directly inside the app.

Apple Intelligence beta requirements iOS 27 macOS 27

The rollout begins as a beta for devices that support Apple Intelligence, which means the software requires iOS 27, iPadOS 27, or macOS 27 and later. That version of the operating system also introduces support for the Model Context Protocol, an open standard that lets third-party AI agents plug into the iWork apps. Developers can now build agents that read and write document data, opening the door for specialized workflows that go beyond what Apple's own assistant provides.

Apple creative apps updates Final Cut Pro Logic Pro

Beyond the core productivity trio, Apple detailed updates for its creative lineup. Pixelmator Pro is receiving a Generate Fill feature that lets users describe an edit, such as removing an object or extending a background, and watches the tool match lighting, perspective, and texture so the change blends naturally. Logic Pro for Mac and iPad gets a Liquid Glass redesign alongside an AI assistant that can set up a vocal track from a single prompt and make edits to existing projects. A redesigned Browser on the Mac version uses on-device intelligence to analyze, classify, and tag imported sounds automatically.

Final Cut Pro is also being updated. The new release pulls templates, graphics, titles, and transitions directly into the timeline through an integration with MotionVFX, and it adds a Content Hub stocked with royalty-free videos, photos, and backgrounds. Apple says the Creator's Studio enhancements will be available to current subscribers at no extra cost starting in November.

Model Context Protocol Apple Intelligence developers

For developers and power users, the most consequential change may be the Model Context Protocol support. By standardizing how external agents connect to Pages, Keynote, and Numbers, Apple is creating a surface where specialized AI tools, whether built for financial modeling, legal drafting, or scientific visualization, can operate inside familiar documents without requiring a separate interface. That could shift the iWork suite from a closed set of features to an extensible platform, especially in enterprise and education settings where custom automation has traditionally relied on scripting or macros.

The timing aligns with Apple's broader strategy of embedding intelligence across the operating system rather than isolating it in a standalone chatbot. Because the assistant runs on devices that meet the Apple Intelligence hardware threshold, many of its operations stay local, which the company has emphasized as a privacy advantage. However, the partnership with OpenAI suggests that some requests may still route to cloud models when the task exceeds on-device capacity, a detail Apple has not fully specified in today's announcement.

Anyone building workflows around the iWork apps should test the beta as soon as it appears on supported hardware. The assistant's ability to manipulate document structure directly, inserting rows, reordering slides, rewriting paragraphs, means existing templates and automation scripts may need adjustment. Teams that rely on third-party plugins should also verify compatibility with the new protocol layer, since agents built on the standard will have read-write access to document content by default.

Creative professionals using Pixelmator Pro, Logic Pro, or Final Cut Pro have a more immediate incentive to upgrade. The Generate Fill tool, the vocal-track assistant, and the integrated Content Hub all ship as part of the standard subscription, so there is no additional purchase required. Editors and musicians who have been waiting for native AI features that respect project context may find these updates reduce the need to round-trip assets through external generative services.

Overall, the announcement signals that Apple views generative AI as a horizontal layer rather than a vertical product. By weaving it into the apps people already open every day, the company is betting that the most valuable use cases emerge when the model understands the document, the timeline, or the session, not just the prompt. Developers who want to reach that context should start experimenting with the Model Context Protocol now, while the beta window is open and the API surface is still stabilizing.

Frequently asked questions

Which apps are getting Apple's new AI assistant and what can it do in each?

The AI assistant is rolling out to Pages, Keynote, and Numbers. In Keynote it designs slides, adds animations, and refines decks. Pages gains drafting, restructuring, and wording tweaks. Numbers can calculate results, visualize trends, organize tables, and recommend chart types based on existing data.

What devices and OS versions support the new AI assistant beta?

The beta requires devices that support Apple Intelligence, meaning iOS 27, iPadOS 27, or macOS 27 and later. Only hardware meeting the Apple Intelligence threshold can run the assistant, with many operations staying on-device for privacy.

What is the Model Context Protocol and why does it matter for developers?

The Model Context Protocol is an open standard introduced in the new OS versions that lets third-party AI agents plug into Pages, Keynote, and Numbers. Developers can build specialized agents for financial modeling, legal drafting, or scientific visualization that read and write document data directly inside the apps.

What new AI features are coming to Pixelmator Pro, Logic Pro, and Final Cut Pro?

Pixelmator Pro gets Generate Fill for object removal and background extension with natural blending. Logic Pro adds a vocal-track assistant from a single prompt and an AI-powered Browser that classifies and tags sounds. Final Cut Pro integrates MotionVFX templates into the timeline and adds a royalty-free Content Hub.

When will the creative app updates be available and will they cost extra?

The Creator's Studio enhancements for Pixelmator Pro, Logic Pro, and Final Cut Pro will be available to current subscribers at no extra cost starting in November. The iWork AI assistant beta rolls out first on supported devices running the new OS versions.

Source: The Verge

models

DeepMind Unveils Genie 4 World Model With Better Physics and Audio

Google DeepMind unveiled Genie 4, a world model generating interactive simulated environments from text descriptions. Powered by Gemini 3.8 Flash for reasoning, it features improved physics, realistic human movement, and native audio generation-including adaptive spoken dialogue-a first for DeepMind. The simulation remains consistent for only a few minutes. Access is limited to trusted testers, developers via Gemini API and AI Studio, and Google AI Ultra subscribers ($100/month), with

models

Google DeepMind Unveils Genie 4 World Model With Native Audio

Google DeepMind unveiled Genie 4 on October 27, 2026, a natively multimodal world model generating audio, music, ambient noise, and dialogue, simultaneously with visual content in real-time interactive simulations. Built on the Gemini model family, it maintains spatial and temporal coherence across extended sessions. The research preview is limited to trusted testers, Google AI Ultra subscribers in Project Genie, and early partners, with a public waitlist open but no timeline for broad

industry

Anthropic Files IPO to Beat OpenAI to Public Markets

Anthropic filed its S-1 registration with the SEC on October 26, planning to list on Nasdaq under ticker ANT and become the first major foundation model provider to go public, moving ahead of rival OpenAI. The filing reveals a $80.7 billion net loss on $16.3 billion revenue in the first half of 2026, largely from accounting charges, and roughly $369 billion in compute commitments over the next decade.

Share this post
Report