AI Experiments 0.5.0 has been released and is available for download! “What’s new in AI Experiments…” posts (labeled with the #aiex-release tag) are posted following every AI Experiments release, showcasing new features included in each release.

We’re pleased to announce the release of AI Experiments v0.5.0, the latest update to the canonical pluginPlugin A plugin is a piece of software containing a group of functions that can be added to a WordPress website. They can extend functionality or add new features to your WordPress websites. WordPress plugins are written in the PHP programming language and integrate seamlessly with WordPress. These can be free in the WordPress.org Plugin Directory https://wordpress.org/plugins/ or can be cost-based plugin from a third-party. powering experimental AI-powered features in WordPress. This release focuses on preparing the plugin for the upcoming WordPress 7.0 release, simplifying dependencies and aligning the plugin with the AI functionality now landing in coreCore Core is the set of software required to run WordPress. The Core Development Team builds WordPress. with the Connectors screen.

What’s new in 0.5.0?

Aligning with WordPress 7.0

One of the major goals of this release is aligning the plugin with WordPress 7.0’s built-in AI capabilities, specifically the shift of the WP AI Client from a standalone, canonical Composer package to bundled in WordPress core.

In previous versions, the AI Experiments pluginAI Experiments Plugin WordPress's AI laboratory bringing all building blocks together. Serves as both a user tool and developer reference implementation. First release (v0.1.0) includes Title Generation experiment. included its own dependencies for AI client functionality. With 0.5.0, those dependencies have been removed and the plugin now relies on WordPress core’s AI infrastructure instead. This helps ensure the experiments remain compatible with the direction of AI development in core and reduces duplication between the plugin and WordPress itself.

As part of this shift, the plugin now sets WordPress 7.0 as its minimum supported version, reflecting the move toward using the AI features provided directly by WordPress which are only available from WordPress 7.0 onward. If you’re using the plugin with WordPress 6.9, then you’ll need to wait for the release of WordPress 7.0 on April 9th to upgrade the plugin (or alternatively can upgrade to a betaBeta A pre-release of software that is given out to a large group of users to trial under real conditions. Beta versions have gone through alpha testing in-house and are generally fairly close in look, feel and function to the final product; however, design changes often occur as part of the process./RCRelease Candidate A beta version of software with the potential to be a final product, which is ready to release unless significant bugs emerge. version of WordPress 7.0 to continue testing).

Improved Connector Integration

The plugin has also been updated to better align with the connector-based architecture used in WordPress for connecting AI providers.

Credentials previously stored through the plugin’s AI Credentials screen are now migrated to the Connectors screen, ensuring compatibility with the broader ecosystem of AI provider plugins and the connector APIs in core.

This change makes it easier for experiments to interact with whichever AI providers a site has configured, while keeping providerProvider An AI service offering models for generation, embeddings, or other capabilities (e.g., Anthropic, Google, OpenAI). management centralized in WordPress.

What’s next in 0.6.0?

Work is already underway on several features and refinements planned for v0.6.0, including:

• Onboarding flows as necessary in support of the “Try AI” callout in WordPress 7.0
• Refining Experiments to potentially elevate to Feature within the plugin, while renaming the plugin from AI Experiments to WordPress AI
• Image Editing experiment to erase and replace, expand canvas and fill, and background removal
• Refine from Notes experiment to update post content based on editorial feedback
• Content Provenance tracking for text and images via C2PA
• Contextual Tagging experiment that suggests post tags and categories based on post content, title, and excerptExcerpt An excerpt is the description of the blog post or page that will by default show on the blog archive page, in search results (SERPs), and on social media. With an SEO plugin, the excerpt may also be in that plugin’s metabox., helping authors apply consistent, relevant taxonomyTaxonomy A taxonomy is a way to group things together. In WordPress, some common taxonomies are category, link, tag, or post format. https://codex.wordpress.org/Taxonomies#Default_Taxonomies. directly in the editor
• Updating AI Experiments admin screens to utilize DataViews and DataForms

Several early prototype experiments are also being explored, including type-ahead suggestions, content moderation assistance, extended provider support, AI request logging, and tools like the AI Playground and deeper MCP integration. These concepts are still exploratory, but they help test how AI could support real workflows across WordPress. We encourage users and developers to review and test these ideas and share feedback so the most valuable experiments can mature and land in upcoming releases like 0.6.0.

Thanks to contributors!

A big thank you to everyone who contributed to this release, including:

@dkotter, @jeffpaul, @isotropic, @flixos90, @juanfra, and others involved in review, testing, and 83 commits between 0.4.0 and 0.5.0.

Your help and feedback are what make these experiments possible.

Get involved

As always, we welcome feedback, testing, and contributions from the community. Whether you are interested in editor UXUX UX is an acronym for User Experience - the way the user uses the UI. Think ‘what they are doing’ and less about how they do it., APIs, accessibilityAccessibility Accessibility (commonly shortened to a11y) refers to the design of products, devices, services, or environments for people with disabilities. The concept of accessible design ensures both “direct access” (i.e. unassisted) and “indirect access” meaning compatibility with a person’s assistive technology (for example, computer screen readers). (https://en.wikipedia.org/wiki/Accessibility), performance, or AI ethics and policy, there are many ways to participate.

You can explore the v0.5.0 release today, review open issues and pull requests, and help shape what comes next.

Props to @justlevine for reviewing this post.

#ai-experiments, #ai-release, #aiex-release, #canonical-plugins, #core-ai