Skip to content

What is Responsible AI? Definition and explanations

Responsible AI is the set of principles and practices aimed at making an AI system useful without causing unjustified harm: fairness across groups of people, transparency about its use, explainability of decisions, data protection, robustness and human oversight. The NIST AI Risk Management Framework offers an operational rather than declarative version of these.

In one sentence

Responsible AI aims at a system that is useful without unjustified harm: fairness, transparency, oversight.

Related terms

Sources and references

Glossary