Artificial Intelligence (AI)
Artificial intelligence (AI) is software that performs tasks that normally require human judgment, such as understanding language, recognizing patterns, making recommendations, or generating content.
For product designers, AI is not a single feature type. It is a capability layer that changes how interfaces handle input, uncertainty, errors, and trust.
What it means
AI systems learn from data and instructions to produce outputs that adapt to context, rather than following only fixed if/then rules.
Why designers should care
When you design AI products, you are designing for probabilistic behavior: users need clarity about what the system can do, when it might fail, and how to recover.
Example
A support copilot suggests reply drafts from past tickets. The UX must show that suggestions are AI-generated, let agents edit before send, and offer a fallback when confidence is low.
Common mistakes
- • Treating AI like a deterministic button that always returns the same result.
- • Hiding AI involvement and surprising users when output quality shifts.
- • Designing only the happy path without empty, error, or “I don’t know” states.