WordPress.com, the hosting platform owned by Automattic, announced Tuesday that it is rolling out a built-in AI assistant for websites. The tool works directly inside a site and understands its content and layout, allowing owners to make changes using natural language commands.
Users can modify layout, design, and other visual patterns by typing instructions to the assistant. Changes appear on the site in real time as edits are made. The company said prompts do not need to be highly specific. General instructions such as asking to make a section feel more modern or spacious, brighten site colors, or suggest cleaner and more professional fonts, are supported.
How to Enable
The WordPress AI Assistant is optional and must be enabled manually. Users can activate it by logging in, opening their Sites list, selecting a site, navigating to Settings, scrolling to AI tools, and toggling on the “Enable AI assistant” option.
Customers who purchase a website through the AI website builder will have the assistant enabled automatically.
Layout and Content Updates
The assistant can also create or adjust structural elements. For example, users can ask it to add a contact page or insert a testimonials section below an existing block. These changes work with block themes only. The AI assistant does not appear in the editor for classic themes.
In addition, the tool can update written content. Site owners can request rewrites of text, such as making a biography sound more confident, or translating sections into another language.
Editing and Collaboration Features
The AI also functions as an editor. It can suggest headlines, check facts, and provide grammar and writing recommendations. This capability is integrated into the block notes editor introduced in WordPress 6.9, which supports collaboration between team members. Users can access the assistant within this workflow by typing @ai followed by a request. The AI responds in the editor and includes relevant links and citations when referencing external sources.
AI Image Tools
For visual content, the assistant uses Google Gemini’s Nano Banana AI models. These models can generate new images or modify existing ones. The feature appears as a “Generate Image” button in the Media Library, where users can define image styles or set aspect ratios.
