Why This Landscape Matters
“AI” is often used as a single label for very different products and features.
In practice, an AI assistant, a predictive model, and an AI search tool don’t behave the same way. They create different expectations, fail in different ways, and require different explanations.
When documentation treats them as interchangeable, users are left guessing. Onboarding becomes vague. Expectations drift.
This page helps you identify what kind of AI system you are documenting and why that classification should shape how you write about it.
Who This Section is For
This page is for:
- Technical writers documenting AI-powered products or features
- Writers unsure how to categorize an AI feature
- Teams who need shared vocabulary before documentation work begins
How to Use This Page
Think of this page as a map.
Each category below links to a dedicated article. Those articles explain how that type of AI system behaves and where users tend to misunderstand it.
If you want a clearer mental model of what AI is and isn’t before classifying products, start with What Artificial Intelligence Is (and Isn’t).
Otherwise, start on this page when you are:
- Scoping documentation for a new AI feature
- Reviewing a product requirements document
- Aligning terminology with product or engineering teams
Core AI Product, Feature, and System Categories
Each category below represents a distinct pattern of user experience. Documentation approaches differ because the behavior differs.
AI Assistants, Copilots, and Agents
Embedded AI features that help users complete tasks within an existing workflow.
Generative AI
AI systems that generate new content such as text, images, audio, or video based on user input.
Conversational AI
Conversation-first systems where dialogue itself is the primary interface and experience.
Predictive AI
AI systems that analyze data to forecast outcomes, risks, or probabilities.
AI Search and Retrieval
AI-powered search systems that retrieve relevant information from knowledge sources using semantic understanding.
AI Workflow Automation
AI systems that trigger, execute, or orchestrate actions across tools and processes.
AI Personalization Systems
AI systems that adapt content, recommendations, or experiences based on user data and behavior.
AI Safety and Guardrails
Systems that constrain AI behavior to ensure safe, compliant, and responsible use.
What Makes AI Systems Agentic?
Understanding how AI systems plan, sequence, and execute actions across tools
Why These Distinctions Matter for Documentation
Different AI categories create different user expectations.
A generative feature raises questions about variability and accuracy. A predictive model raises questions about confidence and scoring. An automation system raises questions about control and reversibility.
If all AI features are documented the same way, important differences disappear. The result is vague documentation, misplaced trust, and underused features.
Clear classification makes clearer writing possible.
How This Page Fits Into the Series
This page helps you classify the AI feature or product you are documenting. It provides shared terminology for the rest of the series.
It does not explain how to document each category. For that, read the individual category articles and the corresponding “Documenting X” guides.
Where to Go Next
- Read the category article that matches your product or feature.
- Then read the corresponding “Documenting X” guide in AI Documentation Guides.
- Use Technical Writing Best Practices for broader writing principles.
AI documentation becomes clearer when you pause to identify what kind of system you are describing. Classification isn’t academic. It shapes how users understand the product.