INTRO TO AI PRODUCT DESIGN

Get an introduction to the course structure, objectives, and capstone project, as well as identify a need/pain point to explore AI solutions.

• Instructor introduction
• Course objectives & flow
• The AI landscape today
• Capstone project 
• Q&A

Dive in! Gain a base understanding of AI, how users experience it, and how it makes its way into product development.

• AI overview & applications
• Current trends
• Human-centered design: AI product development methodologies
• AI product lifecycle
• Case Study: Human-computer interaction of Google Maps

Assignment #1: Critique an existing AI-powered product (app, website, etc.) from a human-centered design perspective. Identify areas where the product excels and where it could improve.

Explore the basics of machine learning, including supervised and unsupervised learning, and get a demonstration of a linear regression-based prediction.

• What is machine learning?
• Algorithms: Supervised & unsupervised 
• Demo: Linear regression-based prediction

Assignment #2: Propose a potential AI solution to a user problem using supervised or unsupervised learning. Write a one-page proposal outlining how AI could address the problem.

Delve into the fundamentals of neural networks, deep learning, and generative AI, as well as how to build a neural network and interact with LLMs for content generation.

• Neural networks & their architecture
• Applications of deep learning
• Intro to gradient descent & backpropagation
• Generative AI and LLMs
• Demo/DIY: Build a neural network & image classification
• Demo/DIY: Interacting with LLMs to generate personalized content

Assignment #3: Design a user flow and wireframe for an AI-powered feature (e.g., personalized recommendations, chatbot) using deep learning, Generative AI, or machine learning principles.

In class 4, you will learn to design user interfaces and interactions for AI products, emphasizing explainability, transparency, and inclusivity in AI design. You will also explore prototyping techniques using Figma.

• User interface (UI) and interaction design for AI products
• Case Study: Visualizing AI-driven recommendations
• Explainability & transparency in AI
• Inclusivity & diversity in AI product design
• Demo: Advanced Figma techniques for prototyping AI interfaces

Explore biases, fairness, and privacy in AI algorithms and data, and discover how to design AI systems that align with societal values and comply with industry regulations like EU AI Act.

• Biases in AI algorithms & data 
• Fairness & inclusivity in AI design
• Protecting user data & data privacy
• Regulations & standards compliance
• Aligning with societal values

Assignment #4: Analyze an AI system for potential ethical concerns (bias, privacy, inclusivity) and propose design adjustments that address those issues. 

Discover what an IA is and learn about the different personas and what they’re trying to make/solve. 

• Information architecture in AI applications
• Organizing & structuring data for AI products
• User personas & scenarios: Mapping user goals & AI features
• Case Study: Information architectures in successful AI products

Apply AI technologies to solve user problems, prototype AI solutions, conduct user testing, and iterate designs based on feedback. Learn to be intentional while designing for AI to ensure a positive impact. 

• Solving user problems with AI technologies
• Rapid prototyping methods
• Validating AI concepts: Interactive prototypes 
• User testing & iteration 
• Workshop: Prototyping & interactions (in Figma) 

Assignment #5: Build a static/interactive prototype (in tool of choice) that includes an AI-powered feature, incorporating both user flows and layouts. Conduct user testing, gather feedback, and iterate based on the findings.

Delve into user flows for AI-powered interactions, focusing on user-centred AI design and its impact on adoption and engagement.

• User flows for AI-powered interactions
• Foundations of user-centered AI design
• Case Study: Impact of UX design on AI product adoption and engagement

In Class 9, you will explore the phases of AI product development, including managing AI models and understanding issues like hallucinations in LLMs, while exploring RAG pipelines.

• Phases of AI product development
• Managing AI models
• Workshop: Managing the end-to-end process for AI product design
• Demo: Understanding hallucinations & controlling outputs in LLMs 
• Workshop: Building a RAG pipeline in a Python Notebook

Bonus Assignment #6: Using ChatGPT, build a functioning notebook yourself.

Explore how to gather and analyze user research data to guide AI product decisions and validate AI-driven changes through methods like A/B testing.

• Methods for gathering and analyzing user research data 
• A/B testing 
• Case Study: Using research data to improve AI products, such Airbnb’s dynamic pricing
• Generative AI in product research: Principles & limitations 

Assignment #7: Do some research on your capstone project so far. What are possible users saying? What do you need to change? 

Learn how to design multi-modal AI systems that integrate different input methods (voice, text, images) and improve accessibility for users with varying abilities.

• Multi-modal AI systems
• Improving accessibility
• Case Studies: Review real-world examples, such as OpenAI ChatGPT, AirBnb Computer Vision, Instagram Explore, and more.
• Design principles for multi-modal systems

Assignment #8: Try and include a multi-modal experience in your capstone project. (e.g., voice, text, image). Ensure the design includes accessibility features such as voice commands, speech-to-text, or screen readers.

Learn to present AI-powered product designs effectively, including gathering feedback, incorporating revisions, and documenting design decisions.

• Design reviews role
• Structuring effective reviews
• Constructive feedback and iteration
• Design documentation & presentation techniques

Assignment #9: Present your final project in a design review format, showcasing your AI-powered product design. Use peer and instructor feedback to make final revisions before submission.