Course Schedule
AI 380: Human-AI Interaction · Fall 2026
This schedule is subject to change. Updates will be announced on Canvas and Discord. Readings are due before the class session listed. All assignment deadlines are at 11:59 PM on the date listed unless otherwise specified.
Lecture Activity / Workshop / Lab Presentation No Class Due Date
Module 1: Human-AI Interaction and User-Centered Design
Foundations of HCI, the role of humans in AI, design thinking, needfinding, and user research. Establishes the vocabulary and frameworks underpinning all human-centered design practice — and situates conversational AI within the broader HCI tradition.
Week 1 — Aug 24 & 26 · Introduction to Human-AI Interaction
Welcome to AI 380. We introduce the field of Human-AI Interaction, survey the landscape of conversational AI systems — chatbots, voice assistants, recommendation engines — and establish why design matters for AI. Course logistics, expectations, and team formation overview.
What is the relationship between humans and machine learning systems? We examine interactive ML, human-in-the-loop design, and the fundamental tension between automation and human control.
Readings Due Aug 28:
Due Aug 28: HW0 — Personal Introduction & Expectations (Canvas Discussion Board)
Week 2 — Aug 31 & Sep 2 · Historical Perspectives & Everyday AI
From early command-line interfaces to today’s large language models — how did we get here? We trace the arc of human-computer interaction, highlighting where AI shifted the design paradigm from direct manipulation to delegated agency.
AI is embedded in the tools we use every day — and often fails us in everyday ways. We examine user-reported problems with intelligent applications and discuss the gulf between what AI promises and what it delivers.
Readings Due Sep 5:
- Dijkstra, E.W. (1978). On the Foolishness of “Natural Language Programming.”.
- Shneiderman, B. & Maes, P. (1997). Direct Manipulation vs. Interface Agents. Interactions.
- When People and Algorithms Meet: User-Reported Problems in Intelligent Everyday Applications. IUI 2019.
Due Sep 4 (Last Day Add/Drop)
Due Sep 5: HW1 — Critical Literature Synthesis & Interface Critique
Week 3 — Sep 7 & 9 · Understanding Users & Personas
How do we learn what users actually need? We cover participant observation, contextual inquiry, and interview techniques. We then translate user research into personas — actionable archetypes that guide design decisions throughout the project lifecycle.
Readings Due Sep 14:
- Klemmer, S. (2013). Participant Observation. 6-min video. HCI MOOC, Stanford.
- Visme. (2021). How to Create Personas. 10.5-min video tutorial.
Week 4 — Sep 14 & 16 · Design Process & Workshop 1
IDEO’s design thinking process: Empathize → Define → Ideate → Prototype → Test. We apply the framework to AI system design and examine BJ Fogg’s Persuasive Technology model — how AI systems nudge, influence, and sometimes manipulate user behavior.
Hands-on team workshop. Using your personas and problem statements, teams practice the first three stages of design thinking: empathy mapping, POV synthesis, and structured ideation using Brainstorming, Chainstorming, and Tweetstorming techniques.
Readings Due Sep 18:
- IDEO. Design Thinking. Online resource. || Tim Brown urges designers to think big
- Brainstorm, Chainstorm, Cheatstorm, Tweetstorm: Ideation Methods..
- Fogg, B.J. The Process of Persuasion. Interaction Design Foundation.
- Dam, R.F. & Siang, T.Y. Get Started with Prototyping. Interaction Design Foundation
Due Sep 18: HW2 — Applied Needfinding & Low-Fidelity Prototype
Key Date Sep 23: Last day to withdraw (partial semester)
Module 2: Human-AI Collaborations
Conversational AI design, human-AI collaboration challenges, dialogue management guidelines, and lab work building Rasa-powered conversational agents.
Week 5 — Sep 21 & 23 · Conversational AI & Lab 1
Introducing the conversational agent as an interaction paradigm. What distinguishes a conversational interface from a GUI? We examine the design space — text-based chatbots, voice assistants, multimodal agents — and discuss the expectations users bring to each type.
Hands-on lab. Students explore human-in-the-loop design by building a simple conversational prototype and experimenting with different levels of automation and user control. Introduces the Rasa development environment.
Readings Due Sep 28:
- Luger, E. & Sellen, A. (2016). “Like Having a Really Bad PA”: The Gulf between User Expectation and Experience of Conversational Agents. CHI 2016.
- Interaction Design Foundation. What Is User Interface Design?
Due Sep 25: GP0 — Team Formation, Charter & Project Journal Setup
Week 6 — Sep 28 & Sep 30 · Challenges & Guidelines for Human-AI Interaction
Why is designing for AI-mediated interactions uniquely difficult? We examine the fundamental challenges: opacity of AI decision-making, miscalibrated user trust, brittleness, and the problem of explainability. Case studies from deployed systems.
Microsoft’s 18 guidelines for Human-AI Interaction provide a practitioner-oriented framework grounded in empirical research. We walk through each guideline, discuss which apply to conversational AI, and identify gaps or conflicts with standard UX heuristics.
Readings Due Oct 5:
- Cai, C.J., et al. (2019). Re-examining Whether, Why, and How Human-AI Interaction Is Uniquely Difficult to Design. CHI 2020.
- Amershi, S., et al. (2019). Guidelines for Human-AI Interaction. CHI 2019.
Due Oct 2: GP1 — Co-Designing & Authoring VR Interventions for Older Adults
Week 7 — Oct 5 & 7 · Conversational AI (1) & Midterm Quiz
Introduction to the conversational AI ecosystem: rule-based systems, retrieval-based chatbots, generative LLMs, and task-oriented dialogue systems. We examine the CALM (Conversational AI with Language Models) paradigm and how Rasa Pro implements it.
In-class midterm quiz covering Modules 1–2 content: HCI fundamentals, design thinking, needfinding methods, conversational AI design principles, and Human-AI Interaction guidelines.
Readings Due Oct 12:
- Human-AI Interaction & GenAI. CHI 2026 Podcast — link on Canvas.
- Freed, A.R. (2021). Conversational AI. Chapter 1. PDF on Canvas.
Key Date Oct 9: Mid-semester break — No class
Week 8 — Oct 12 & 14 · Conversational AI (2) & Workshop 2
What separates good conversational AI design from bad? We examine dialogue management patterns — happy paths, fallbacks, clarification loops, and graceful degradation. Case studies from commercial assistants (Alexa, Google Assistant) and enterprise chatbots.
Applying design thinking to conversational AI. Teams define their assistant’s scope, ideate dialogue flows, and produce paper-based conversation flow diagrams. Discussion of ethical responsibilities for AI designers and accessibility considerations per WCAG.
Readings Due Oct 19:
- Weiser, M. (1991). The Technologist’s Responsibilities and Social Change..
- W3C. Web Content Accessibility Guidelines (WCAG) 2.1.
Due Oct 16: GP2 — User Needfinding, Personas & Paper Conversational Workflow
Module 3: AI Design and Conversational Orchestration
Deep-dive into building AI assistants with Rasa CALM — from architecture through training, testing, and deployment. Three hands-on lab sessions guide teams through implementation of their conversational AI assistant.
Week 9 — Oct 19 & 21 · CALM Architecture & Lab 2
Deep dive into Rasa Pro’s CALM (Conversational AI with Language Models) architecture: flows, commands, actions, domain, and slot management. How CALM differs from traditional intent-based NLU and why it produces more robust multi-turn dialogue.
Hands-on lab: teams set up the Rasa development environment, implement their first CALM flow, define custom actions, and test basic multi-turn dialogue. Students begin mapping their GP2 conversational workflows into Rasa YAML configs.
Readings Due Oct 26:
Key Date Oct 23: Last day to withdraw from full-semester class
Week 10 — Oct 26 & 28 · CALM Overview & Lab 3
Comprehensive review of CALM components: intent classification, entity extraction, flow conditions, slot filling, and action server integration. Discussion of common failure modes and how to debug Rasa dialogue predictions using the Rasa Inspector.
Lab continuation: teams implement custom actions (API calls, database lookups), add entity-based slot filling, and test fallback handling. Instructor and TAs provide hands-on debugging support.
Readings Due Nov 2:
Week 11 — Nov 2 & 4 · Building for Success & Lab 4
What makes an AI assistant succeed with real users? We examine best practices for intent coverage, response quality, graceful error handling, and inclusive design. Case studies from deployed elder-care and accessibility-focused AI systems.
Final Rasa lab: teams complete their full dialogue system, integrate the web/voice interface, and conduct internal QA testing. Focus on accessibility features (large UI targets, speech-to-text integration, high-contrast styling) and containerizing with Docker for GP3.
Readings Due Nov 9:
Key Date Nov 3: Only lab-format classes meet (check Canvas for schedule)
Module 4: User Experience, Usability Study, Report & Presentation
Usability testing, heuristic evaluation, ethical AI design, peer review, and final project presentations. Students evaluate their systems with real older adult users and communicate findings professionally.
Week 12 — Nov 9 & 11 · Usability Testing & Heuristic Evaluation
How do you evaluate a conversational AI system? We cover intent accuracy metrics, conversation quality metrics, task completion rates, and user satisfaction scales. Discussion of mixed-methods evaluation: combining quantitative logs with qualitative think-aloud data.
In-class workshop: peer heuristic evaluation of each team’s conversational AI system. Teams apply Nielsen’s 10 heuristics to a classmate’s prototype and produce severity-rated findings. Discussion of how to run user studies with older adult participants.
Readings Due Nov 13:
- Freed, A.R. (2021). Conversational AI. Chapter 11. PDF on Canvas.
- Nielsen, J. (1994). How to Conduct a Heuristic Evaluation. Nielsen Norman Group.
- Klemmer, S. Heuristic Evaluation Lecture..
Due Nov 13: HW3 — Heuristic Evaluation Report
Week 13 — Nov 16 & 18 · Interface Testing & Documentation
In-class share-out: teams present preliminary findings from field user tests with older adults (task completion rates, usability friction, key observations, think-aloud excerpts). Instructor and peers provide feedback on findings interpretation and GP4 report direction.
What makes a great HCI project report? Academic writing conventions, figure/table design, and telling a coherent design story. How to give an effective technical presentation with a live demo. Teams outline their GP4 reports.
Readings Due Nov 23:
Due Nov 20: GP3 — System Implementation, Integration & Containerization
Key Date Nov 17: Spring 2027 registration begins
Week 14 — Nov 23 & 25 · Peer Review & Fall Break
Structured peer-review session: teams share an executive summary and prototype demo. Reviewers provide structured written feedback using the GP4 rubric. Discussion of emerging AI interfaces — wearables, ambient intelligence, and embodied AI — and what “Human-AI Interaction” will mean when the interface disappears.
Readings Due Nov 30:
- Eagleman, D. (2015). Can We Create New Senses for Humans? TED Talk (video).
Key Date Nov 25–27: Fall Break — University closed
Week 15 — Nov 30 & Dec 2 · Final Presentations
Public project showcase — first half of teams present. Each team delivers a 12-minute presentation + 5-minute Q&A demonstrating their completed conversational AI system, summarizing the design process and evaluation findings from field testing with older adults.
Remaining teams present. Followed by course-wide reflection: what did we learn? Course synthesis and guidance for the Final Reflection Essay. Key date: Dec 4 is the last day of class.
Due Dec 2: Final Presentation and Course Reflection Essay
Due Dec 3: GP4 — Public Deployment, Usability Testing & Final Documentation
Key Date Dec 4: Last day of class
Key Dates Summary
| Date | Event |
|---|---|
| Aug 24 | First Day of Class |
| Aug 28 | HW0 Due (Canvas Discussion) |
| Sep 2 | No Show Roster |
| Sep 4 | Last Day Add/Drop |
| Sep 5 | HW1 Due |
| Sep 7 | Labor Day — No Class |
| Sep 9 | Classes resume |
| Sep 18 | HW2 Due |
| Sep 23 | Last Day to Withdraw (partial semester) |
| Sep 25 | GP0 Due — Team Charter & Website |
| Oct 2 | GP1 Due — VR Interventions |
| Oct 7 | Midterm Quiz (in class) |
| Oct 9 | Mid-semester break — No Class |
| Oct 16 | GP2 Due — Needfinding, Personas, Workflow |
| Oct 23 | Last Day to Withdraw (full semester) |
| Nov 3 | Lab-only class format |
| Nov 13 | HW3 Due |
| Nov 17 | SP27 Registration Begins |
| Nov 20 | GP3 Due — System Implementation |
| Nov 25–27 | Fall Break — No Class |
| Nov 30 & Dec 2 | Final Project Presentations |
| Dec 2 | Final Reflection Essay Due |
| Dec 3 | GP4 Due — Public Deployment & Final Docs |
| Dec 4 | Last Day of Class |