UX and AI Digest Episode 5: Managing Users' Expectations with AI

UXChange Podcast on YouTubeWatch on YouTube

SaveMark complete

Sign in to save videos and track what you’ve finished.

IntermediateAI20 min

About this video

A UXChange podcast episode on research into cognitive surrender, where people follow AI advice even when it is wrong, and on Stanford findings about sycophantic AI models. The host argues that AI products need design that interrupts blind trust rather than static disclaimers.

What you’ll learn

  • Explain cognitive surrender and why users follow incorrect AI output
  • Assess how sycophantic AI behaviour can reinforce misplaced trust
  • Consider design approaches that interrupt blind trust in AI output