I had a client reach out last week—a high-traffic publishing site—and their admin area looked like a graveyard of “AI-powered” plugins. They had one for title suggestions, one for alt text, and another for content summaries. Each one had its own API key, its own settings page, and they were all fighting for the same hook in the block editor. Total nightmare. The site was sluggish, and the editors were constantly complaining about UI conflicts. Standardized AI features? They were non-existent.
My first instinct was to just write a custom wrapper for them. I thought I’d just pipe everything through a single OpenAI endpoint and be done with it. But honestly, that was a mistake on my part. I would have just been adding a fourth “black box” to their stack. The real fix wasn’t more custom code; it was moving toward a standardized framework. That’s exactly what the WordPress AI team is aiming for with their new release.
Standardized AI Features: A Real Framework for WordPress
The release of the AI Experiments plugin (v0.1.0) is a massive step in the right direction. It’s not just another plugin; it’s a “canonical” space for the community to figure out how AI should actually live in core. Instead of everyone inventing their own way to talk to LLMs, this plugin introduces a Registry and Loader system. It’s based on the idea that AI should be a set of “Abilities” that any plugin can tap into. I saw the announcement over at the WordPress AI blog and it clicked—this is the “WordPress way” to handle AI.
Here’s the kicker: they’ve included an abstract base class for experiments. This means if you want to build a feature—say, generating a product description—you don’t have to worry about the connection logic or the UI for API keys. You just extend the base class and register it. Trust me on this, this kind of consistency is what saves us from technical debt three years down the line.
/**
* Example of how we might register a standardized experiment
* using the patterns introduced in v0.1.0.
*/
function bbioon_register_product_advice_experiment() {
if ( ! class_exists( 'WordPress\AI\Experiments\Registry' ) ) {
return;
}
$registry = \WordPress\AI\Experiments\Registry::get_instance();
// Registering our custom logic under the new standard
$registry->register(
'bbioon-product-advice',
[
'title' => __( 'AI Product Advisor', 'bbioon' ),
'description' => __( 'Suggests upsells based on product specs.', 'bbioon' ),
'enabled' => true,
]
);
}
add_action( 'wp_ai_experiments_init', 'bbioon_register_product_advice_experiment' );The first experiment in this release is Title Generation. It’s simple, but it demonstrates the WP AI Client SDK in action. It suggests alternative titles directly in the editor, but it does it using a standard API that can eventually support multiple providers. No more hardcoded OpenAI calls in your functions.php. Period.
So, What’s the Point?
If you’re building AI features today, stop doing it in a vacuum. The 0.1.0 release is the foundation for things like Image Generation, Alt Text suggestions, and even an “Abilities Explorer” that’s coming in v0.2.0. By following these patterns now, you’re ensuring your site won’t break when these features eventually land in WordPress core. It’s about being pragmatic, not just chasing the hype.
Look, this stuff gets complicated fast. If you’re tired of debugging someone else’s mess and just want your site to work with modern standards, drop my team a line. We’ve probably seen it before.
Are you still building custom OpenAI wrappers for every little task, or are you looking at these core-led standards yet?
Leave a Reply