Home / Insights / Glossary / Responsible AI

Glossary

Responsible AI

Governing AI systems so their use is accountable, explainable and reviewable.

What it means

Responsible AI covers the governance around an AI system rather than the model itself: what it is approved to do, what evidence supported that approval, who owns it, what human oversight applies, and what would trigger a review.

Most organisations meet it first as a single concrete decision — approving one tool or one feature — rather than as a programme. Starting with an inventory of everything is the common way to spend a year without approving anything.

Where MyRISK fits

What we do about it

MyRISK treats an AI approval as a decision to be recorded like any other high-stakes decision: the boundary, the evidence, the authority and the review trigger. We do not sell an AI governance inventory.

MyRISK Trace

Is this the thing you are actually trying to fix?

A definition rarely settles it. Tell us what happened — the request, the finding, the challenge or the incident — and we will say where to start, or that it isn't us.