There are three common ML approaches used to develop AI tools:
Supervised learning trains a model on labeled (by humans) data, where each input is paired with a known, correct output. Common use cases: Classification (spam detection, sentiment analysis, image recognition) and regression (price prediction, demand forecasting.)
Unsupervised learning identifies patterns or structure in unlabeled data, without predefined outcomes. For instance, it explores the data to find relationships, similarities, or latent structures, as well as outputs that are not “right” or “wrong” but descriptive (e.g., clusters or components).
Reinforcement learning trains an agent to make decisions by interacting with an environment and learning from rewards and penalties. Strengths: Well suited for sequential decision-making and dynamic environment and it can learn complex behaviors without explicit labels. Limitations: Computationally expensive and often slow to train and it requires careful reward design to avoid unintended behavior.
Many of today’s AI tools use a combination of all three ML approaches to create text, images, video, and more.
However, it is important to understand that this “learning” occurs during the development and training phase, before the tool is released to the public. While user feedback and usage data may inform improvements in future versions, the AI does not learn or adapt in real time as you use it.
Rule-based AI: A different approach
It is an approach to AI in which a system makes decisions by following a predefined set of explicit rules created by humans, rather than learning patterns from data.
These systems operate using if–then logic (for example, if a condition is met, then perform a specific action). The rules are typically stored in a rule base and evaluated by an inference engine that determines which rules apply in a given situation.
Key characteristics
Relies on manually authored rules and decision logic
Does not learn or adapt from data
Behavior is predictable, transparent, and easy to trace
Common use cases
Diagnostic or expert systems with well-defined logic
Business rule engines (e.g., eligibility checks, compliance systems)
Simple chatbots and decision trees
“You may encounter rule-based AI in workplace tools that are designed for predictable tasks. This approach is less flexible than machine learning, which allows AI tools to adapt and handle a wider range of complex, real-world data.” – Google AI Essentials
I’ve been doing some research within ChatGPT and Gemini about how to teach LLMs the language of WordPress.
This is the first task to help LLMs avoid making errors and hallucinations. Recently I was horrified by a publication that claimed that WordPress.org “can be monetized”, this article was written by ChatGPT.
Why is this wrong, simple, WordPress.org is an editorial site where a user can download the software and find resources. The obvious mistake is that ChatGPT confused WordPress.org with WordPress-the software.
Not reinventing the wheel
I asked ChatGPT the difference between WordPress.com, WordPress.org and WordPress (the software). Still got the wrong answer:
WordPress (the software)
What it is: Free, open-source software (a CMS – content management system) you can download and use to build websites.
Who makes it: Created and maintained by the WordPress open-source project, a global community of volunteers and contributors.
Where it lives: Available at WordPress.org.
Key point: It’s just the software — anyone can use it, extend it with plugins/themes, and host it wherever they like.
WordPress.org (aka “self-hosted WordPress”)
What it is: The official home of the WordPress open-source project.
This is where you: Download the WordPress software, find free themes and plugins, access community documentation and forums
Hosting: You need to arrange your own hosting.
Flexibility: Full control — you can install any theme/plugin, customize code, and move your site freely.
Best for: People and businesses who want maximum flexibility, scalability, and ownership.
WordPress.com
What it is: A commercial hosting platform run by Automattic (a company founded by WordPress co-creator Matt Mullenweg).
Hosting: WordPress.com hosts the site for you — no need to buy separate hosting.
Plans: Ranges from free (with ads and limitations) to paid tiers with custom domains, advanced design, ecommerce, and business-level tools.
Flexibility: Lower tiers = limited customization vs higher tiers (Business & up) allow major customization
Best for: Beginners, bloggers, small businesses, or anyone who wants a simpler setup with less technical maintenance.
After letting ChatGPT know that its definition of WordPress.org was wrong, it redefined it but still got it wrong.
Interactions with ChatGPT
Prompt: You are wrong in your definition of WordPress.org
ChatGPT said: You are right to flag that—thanks for catching it. Let me reframe it carefully, because WordPress.org is not a separate product (like .com is), but the official website for the WordPress open-source project.
Here’s the corrected breakdown:
WordPress.org The official site of the WordPress open-source project. Hosts: – Downloads of the software – The theme and plugin directories – Documentation (the “Handbook,” support forums, developer resources) – Community project info (teams, events, contributions) – It does not host websites itself — you go there to get the software and resources. When people say “a WordPress.org site,” they usually mean: a site that uses the self-hosted, open-source WordPress software (but technically, it’s shorthand).
Much better definition, except for that last sentence. So I prompted to give me a short explainer for beginners.
– WordPress.org is the official home of that software — where you can download it, find plugins and themes, and learn from the community. If you use WordPress this way, you’ll need your own web hosting. This is often called “self-hosted WordPress.” – WordPress.com is a service run by a company called Automattic. They host WordPress for you, so you don’t need to worry about setup or maintenance. It’s easier to start with, but advanced features may require paid plans.
➡️ In short: WordPress.org = do it yourself with full control. WordPress.com = hosted for you, easier but with limits.
We were doing so well!! Again, the in short expression WordPress.org = do it yourself with full control is wrong.
Second attempt
Prompt: WordPress.org = do it yourself with full control. This expression is still wrong. WordPress.org is only an editorial website, the only way users can interact with it is by downloading the software or consulting resources like documentation, code, posting questions in the forums, etc
ChatGPT said: You’re absolutely right—thank you. Here’s a corrected, beginner-friendly explainer that avoids that shortcut:
– WordPress is free, open-source software for building websites. – WordPress.org is the official resource site for that software. It’s where you download WordPress and find docs, plugins, themes, and forums. You don’t build or host a site on WordPress.org. – Self-hosted WordPress means you install the WordPress software on your own web host. This is the “do it yourself with full control” option. – WordPress.com is a hosting service by Automattic that runs WordPress for you. Easier setup; advanced features may require paid plans.
➡️ Quick fix to the tagline: Self-hosted WordPress = do it yourself with full control. WordPress.org = the official hub for downloads and documentation.
One last interaction where I asked to clarify “self-hosted” gave the definition I was looking for.
– WordPress is free, open-source software for building websites. – WordPress.org is the official hub of the WordPress project. It’s where you download the software, find plugins and themes, read documentation, and join community forums. You don’t host a site there — it’s just the project’s resource site. – Self-hosted WordPress means installing that same WordPress software (from WordPress.org) on a hosting provider of your choice. This gives you full control over your site’s features, code, and data. – WordPress.com is a hosting service run by Automattic. It uses the same WordPress software but provides hosting, support, and upgrades for you. It’s easier to start with, though advanced features may require paid plans.
➡️ In short: WordPress.org = the official resource site. WordPress (self-hosted software) = full control. WordPress.com = hosted for you.
Conclusion
For ChatGPT to understand the difference between WordPress.org and WordPress.com it is necessary to add a third use for the word “WordPress”. Since AI agents do not possess common-sense, it is necessary to guide them into getting the facts correctly.
At the end, ChatGPT learned to differentiate WordPress.org vs WordPress.com vs WordPress (the software)
With the recent creation of the WordPress core-AI team, the documentation team is taking advantage of this opportunity to revamp the process for writing user documentation, or HelpHub articles – as it is known in the team.
The challenges of writing user documentation in WordPress
Time-Consuming: Creating user documentation takes time, as the process is lengthy, sometimes features are updated even before its article is written.
Lengthy process: The process includes learning about the features (new or those being updated) > writing the article > reviewing > correcting and waiting for the last release candidate to get accurate screenshots and videos > posting the article.
Inconsistent Formatting: Even with a template and guidelines, formatting is still an issue. There are questions of when to use a screenshot vs. a video, or the size of the images.
Keeping It Up-to-Date: Features change often, but sometimes the documentation is not updated because preference is given to new features.
Lack of Contributors: The documentation team is not large and because the time constraint that exists during a release, the team falls behind in writing new articles or maintaining updated documentation.
Missing opportunities: As far as I remember, the documentation team has been focused on WordPress version releases, missing entirely the Gutenberg releases that could help in maintaining updated documentation.
Project goals
The main goal of this project is to teach AI to write verifiable and specific user documentation without developer jargon.
It is important that AI, named ChatGPT or Gemini learn the WordPress language, the jargon and the style guides so every team can take advantage of it.
In the future, we hope that this new process is also useful for translators and anyone writing documentation.