Designing AI Experiences
Interfaces for Systems That Can Be Wrong
- Pages
- 81
- Chapters
- 68
- Level
- Beginner–Intermediate
- Format
- E-book
About this book
An AI feature succeeds or fails in its interface. Users need to understand what it can do, judge when to trust it, and recover easily when it is wrong. This book shows designers how to make that happen.
It covers onboarding and expectation-setting, input design and prompting interfaces, presenting outputs with sources and uncertainty, error handling and correction, feedback loops, conversational design, agents that take actions, accessibility, and research methods for AI products. Each chapter includes patterns, anti-patterns and checklists.
The book is tool-agnostic and suitable for designers, researchers and product teams.
What you will learn
- Setting expectations and onboarding users
- Designing inputs and prompting interfaces
- Showing sources, confidence and uncertainty
- Designing for errors, correction and recovery
- Conversational and agentic interfaces
- Accessibility and researching AI experiences
Table of contents
- IntroductionIntroduction
- Chapter 1Why AI Changes Design
- Chapter 2The User's Mental Model
- Chapter 3Principles for AI Experiences
- Chapter 4Automation Versus Augmentation
- Chapter 5Matching Interface to Confidence
- Chapter 6The Cost of Errors in Design
- Chapter 7Avoiding Anthropomorphism Traps
- Chapter 8Consistency and Predictability
- Chapter 9Ethics as Design Material
- Chapter 10Onboarding
- Chapter 11Communicating Capabilities and Limits
- Chapter 12Empty States and First Prompts
- Chapter 13Input Design
- Chapter 14Prompting Interfaces
- Chapter 15Context Awareness
- Chapter 16Clarifying Questions
- Chapter 17Multimodal Input
- Chapter 18Input Privacy Cues
- Chapter 19Templates and Presets
- Chapter 20Inline Versus Separate AI
- Chapter 21Presenting Outputs
- Chapter 22Showing Sources
- Chapter 23Communicating Uncertainty
- Chapter 24Multiple Options
- Chapter 25Explanations
- Chapter 26Error States
- Chapter 27Correction Flows
- Chapter 28Undo and Reversibility
- Chapter 29Confirmation for High-Stakes Actions
- Chapter 30Graceful Degradation
- Chapter 31Latency Design
- Chapter 32Hallucination-Aware Design
- Chapter 33Over-Reliance
- Chapter 34Conversational Design
- Chapter 35Personality and Tone
- Chapter 36Memory and Continuity
- Chapter 37Handoff to Humans
- Chapter 38Agents That Act
- Chapter 39Permissions for Agents
- Chapter 40Activity Logs
- Chapter 41Personalisation
- Chapter 42Notifications From AI
- Chapter 43Feedback Design
- Chapter 44Closing the Feedback Loop
- Chapter 45Collaborative AI in Teams
- Chapter 46Attribution and Labelling
- Chapter 47Accessibility
- Chapter 48Inclusive Design
- Chapter 49Dark Patterns to Avoid
- Chapter 50Research Methods for AI Products
- Chapter 51Wizard-of-Oz Prototyping
- Chapter 52Testing With Imperfect Outputs
- Chapter 53Measuring Experience Quality
- Chapter 54Design Systems for AI
- Chapter 55Writing for AI Interfaces
- Chapter 56Collaborating With ML Teams
- Chapter 57A Worked Example: Designing a Summary Feature
- Chapter 58Designing for Different Expertise Levels
- Chapter 59Voice Interfaces
- Chapter 60Designing for Children and Vulnerable Users
- Chapter 61Cross-Cultural Design
- Chapter 62Case Study: An Email Assistant
- Chapter 63Case Study: A Research Tool
- Chapter 64Case Study: A Customer Service Bot
- Chapter 65Case Study: A Scheduling Agent
- Chapter 66Checklist: Discovery
- Chapter 67Checklist: Design
- Chapter 68Checklist: After Launch
- AppendixPattern Library Summary
- GlossaryGlossary
Who this book is for
UX and product designers, researchers, writers and product managers working on AI features.
Details
| Format | PDF + web edition |
|---|---|
| Length | 81 pages (approx.) |
| Chapters | 68 |
| Level | Beginner to Intermediate |
| Language | English |
| Topic | Practical AI |
| Author | Latent Library Editorial |
Keep reading