Calibrated Trust

An HCI and human-factors concept, articulated by Lee and See, describing the alignment between a user's trust in an automated or AI system and the system's actual capability in a given context: trusting the system when it is reliable and being skeptical when it is not. Designing for calibrated trust means surfacing meaningful uncertainty signals (confidence scores, fallback states, explanations) so users neither over-rely on faulty outputs nor disengage from useful ones. In accessibility, calibrated trust matters because AI assistive tools - automatic captioning, audio description, scene description, sound recognition - are often deployed in high-stakes settings where users cannot independently sanity-check the output, and confidently wrong AI is harder for disabled users to detect or contest than for sighted or hearing users.