A Practitioner’s Guide to Opening the AI Black Box for Humans
summary
Argues that there are no true black boxes and promotes user‑centric XAI that prioritizes technical accuracy plus human meaningfulness. Demonstrates concrete tools: feature‑importance methods and counterfactual explanations tuned with boundary constraints, plus textual outputs for a real‑estate price estimator. Reports an empirical result: 60% preferred longer explanations in a user study. Builds a prototype that generates realistic counterfactuals and validates them with constructs from communications research to evaluate meaningfulness with user studies.
AI-written summary · about
chapters (7)
- 0:00Explainable AI intro
- 1:51Three truths on AI systems
- 5:42User-centric explainable AI
- 9:55What makes a good explanation
- 16:56Evaluating explanation meaning
- 20:09Real estate pricing example
- 25:08Summary & Q&A
related talks
39mThe Search for Transparency and Accountability in the Age of...
15mIntroduction to Explainable AI
30mTrustworthy AI - opening up the black box
26mUsing Explainable AI to garner trust and overcome adoption inhibitors
41mIf We Want AI to be Interpretable, We Need to Measure Interpretability
48mExplainable Artificial Intelligence (XAI): Why, When, and How?
47mML: Explainable Artificial Intelligence explained- 39mSponsored Session: The Search for Transparency and Accountability in the Age of AI: RAI & XAI as Essential Tools