Behind the AI Curtain - Designing for Machine Learning Products
summary
Describes redesigning a B2B predictive analytics UI that reports bill passage likelihood (users confuse a distribution chart for a time series and misinterpret a 95.6% score). Teams cut numeric clutter, replace complex charts with plain-language summaries showing the top three contributing factors, and test until every user can explain this to a new coworker. Evidence: 140 internal docs reveal confusion; outcome influences pricing because customers who don’t understand scores threaten significant revenue.
AI-written summary · about
chapters (8)
- 0:00Introduction and welcome
- 1:12Why design for AI products
- 9:24Principle one: Less is often more
- 11:18Principle two: Ask the right questions
- 15:59Principle three: Writing well matters
- 18:29Case study: Predictive analytics redesign
- 27:35Future of AI and human collaboration
- 28:17Q&A session
related talks
33mBehind the AI Curtain - Designing for Trust in Machine Learning Products
23mDelivering business success with AI
39mChallenges Applying Traditional UI/UX Principles to M.L.
01mAI Is About to Kill the User Interface
32mExplainable AI Explained for Developers
38mAI in action: How to use practically in iGaming beyond the hype?
25mTranslating AI Capabilities Into Better Business Results
47mMastering Power BI Design with User-Centric Frameworks