Explainable AI (XAI)
AI & Emerging Tech
AI whose decisions can be understood and questioned by a person.
In practice
Explainable AI covers the methods and design choices that make a model's output interpretable — why it recommended this, what it weighed, how sure it is. It is partly a technical field and partly an interface problem: an explanation nobody can read is not an explanation. For anything affecting people's money, health or rights, it is not optional.
Questions
What is Explainable AI (XAI)?
Explainable AI is the practice of making a model's reasoning and confidence legible, so a person can understand, trust or challenge what it decided.
Why do you use Explainable AI (XAI)?
Because a decision you cannot question is one you cannot correct, and people rightly distrust systems that act on them without saying why.
How do you use/apply Explainable AI (XAI)?
Show the factors that mattered, in plain language, at the moment of the decision. Expose confidence honestly. Give a path to contest it.
When do you use Explainable AI (XAI)?
Always for high-stakes decisions. For low-stakes ones, at least on request.
Who uses Explainable AI (XAI)?
Designers and teams working on anything that scores, ranks, approves or recommends.
