For designers and product teams building AI features: design interfaces that set accurate expectations, show their work, recover from errors and keep people in control, from chat to agents.
Level
Intermediate
Study time
12 hours
Lessons
19 in 6 modules
Exam
20 questions
About this course
Generative AI breaks familiar interface assumptions: the same input gives different outputs, answers can be fluent and wrong, and agents take actions on a user's behalf. Good AI UX makes these properties visible and manageable instead of hiding them. This course builds on published research-based guidance, including Microsoft's Guidelines for Human-AI Interaction (HAX), Google's People + AI Guidebook and Apple's Human Interface Guidelines for generative AI, and turns it into patterns you can use in design reviews.
Modules cover choosing the right surface (chat, inline assistance, a command bar or a background agent), designing input with prompt scaffolds and examples, presenting output with streaming, citations and appropriate confidence cues, recovering gracefully from errors and refusals, and designing approvals and activity logs for agents. You also learn the transparency requirements that now shape interfaces, including EU AI Act Article 50 disclosure and labeling duties and C2PA Content Credentials, plus accessibility under WCAG 2.2 and how to usability-test non-deterministic features. Facts are stated as of October 2026. The course ends in an automated exam; passing earns a Wirewalk Institute of AI certificate of completion and assessment.
What you will be able to do
Apply the Microsoft HAX guidelines and Google People + AI Guidebook to critique an AI feature design
Choose between chat, inline, command and background-agent surfaces for a given task
Design onboarding and empty states that set accurate expectations of capability and limits
Design output presentation with streaming, citations and calibrated uncertainty cues
Design error, refusal and correction flows that keep users productive
Design approval, undo and activity-log patterns for agents acting on a user's behalf
Meet AI disclosure, content-labeling and WCAG 2.2 accessibility requirements in an AI interface
Plan usability tests for features whose output varies from run to run
Syllabus
6 modules · 19 lessons · about 12 hours
1
Module 1: Principles for Human-AI Interaction
The established guidance and the mental-model problem it addresses.3 lessons
Why AI Breaks Interface Assumptions: Non-Determinism, Fluency and Agency
Microsoft's HAX Guidelines and Google's People + AI Guidebook in Practice
Mental Models and Calibrated Trust: Avoiding Both Overreliance and Disuse
2
Module 2: Choosing the Surface and Shaping Input
Where AI appears in the product and how users tell it what they want.3 lessons
Chat, Inline Assist, Command Bar or Background Agent: A Surface Decision Guide
Onboarding, Empty States and Example Prompts That Set Expectations
Structured Input: Prompt Scaffolds, Context Pickers and Attachments
3
Module 3: Presenting AI Output
Making generated content readable, checkable and honest about uncertainty.4 lessons
Streaming, Skeletons and Progress: Designing for Latency
Citations, Source Previews and Show-Your-Work Patterns
Signaling Uncertainty Without Fake Precision
Multiple Drafts, Variations and Regenerate Controls
4
Module 4: Errors, Refusals and User Control
Designing for the moments when the AI is wrong or cannot help.3 lessons
Designing for Wrong Answers: Edit, Correct and Report Flows
Refusals and Limits: Explaining What the System Will Not Do
Feedback Controls That Produce Useful Signal
5
Module 5: Designing for Agents
Interfaces for systems that plan and act across tools.3 lessons
Plans, Previews and Approvals Before an Agent Acts
Activity Logs, Undo and Interruptibility
Memory and Personalization: Showing and Editing What the System Remembers
6
Module 6: Transparency, Accessibility and Testing
Disclosure obligations, inclusive design and research methods for AI features.3 lessons
AI Disclosure and Labeling: EU AI Act Article 50 and C2PA Content Credentials
Accessible AI Interfaces: WCAG 2.2, Screen Readers and Streaming Text
Usability Testing Non-Deterministic Features
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 40 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by Google, Microsoft, Apple, Adobe, OpenAI, or Anthropic. Product and company names are trademarks of their respective owners.
The product manager's playbook for AI features. You learn to screen use cases for AI fit, run discovery with real model output, write a spec that defines quality with an eval set, choose between APIs, open-weight models and fine-tuning, and plan a staged launch with risk, compliance and support ready on day one.
How to measure AI features end to end. You build a metric tree from business outcome to model quality, instrument implicit and explicit feedback, link offline eval scores to online behavior, and design A/B tests for non-deterministic features with guardrail metrics, variance reduction and sequential testing.
An expert course on the economics of AI products. You build a cost-to-serve model from tokens, caching, retrieval, tools and infrastructure, compare seat, usage, credit, hybrid and outcome-based pricing using public examples, design packaging and limits that protect margin, and implement metering and billing with tools such as Stripe Billing, Metronome and Orb.