Artificial Intelligence

Generative artificial intelligence is the most important new information technology since the World Wide Web. I'm exploring ways to help organizations and individuals create value using AI. I'm sharing what I learn here.

Workshops and Presentations

AI Hands-on: Beyond Chat
Learn to tap the power of generative AI beyond the chat box. Waitlist →

AI as a Thought Partner
A Rosenfeld Media seminar on using AI for personal knowledge management.

Projects and Experiments

Using Graph RAG for Information Architecture
How to use knowledge graphs with retrieval augmented generation (RAG) to organize large content repositories.

LLM-powered Content Tagging
An approach to re-categorizing large-scale content repositories, such as websites, using AI.

Using Embeddings to Group Content
An approach to finding patterns in content by having an LLM create an embeddings database.

LLMapper
An open source command-line tool that generates simple concept maps from Wikipedia pages.

---

Essays

• LLMs as Robot Arms

  How to deliver value by rearchitecting knowledge pipelines around AI constraints.

• Transcribe Handwriting Using a Local LLM

  How to turn handwritten notes into Markdown — on your computer.

• Tapping Into AI’s Conservatism

  AI favors established frameworks. Knowing that will change how you develop your information systems.

• Two Kinds of Writing

  Writing can be a means for thinking. AI can help — but make sure you’re driving.

• AI “Real Talk”: How Useful are AI Agents Right Now?

  An informal livestreamed conversation about a key AI concept.

• Using Claude Code for Information Architecture

  Repurposing a software development agent to help draft website taxonomies.

• Automate the Boring

  Delegating tedious tasks to AI is the most strategic, feasible, and humane use of the technology today.

• Introducing unfinishe 

  Announcing a new strategic consultancy focused on helping small businesses make the most of AI.

• Local GraphRAG: A Progress Report

  Good and bad news on my ongoing experiments with locally-hosted GraphRAG.

• Using AI to Illuminate My Blind Spots

  ChatGPT helps me think about what I need to improve.

• Smarter AI Begins With Your Business’ Knowledge Pipeline

  Making better use of AI starts by understanding how information flows through your business.

• Don't Let the Drawings Distract You: AI is Not About the Output

  AI’s real value isn’t in generating flashy outputs; it’s in making sense of unstructured information to support better decisions.

• AI is Probabilistic – That's Why It Needs Constraints

  Computers used to exclusively follow rules; now they generate possibilities. Combining both approaches maximizes their potential.

• AI and Taxonomies: Creating vs. Applying

  AI can help apply taxonomies at scale, but defining them is still (mostly) a human task.

• Why Website Taxonomies Drift (and What to Do about It)

  Taxonomies drift without proactive attention, making websites harder to navigate and manage. AI can help.

• The New Leverage: AI and the Power of Small Teams

  Doing more with less isn’t just possible — it’s the future of work.

• Book Notes: “A Thousand Brains”

  An intriguing new theory of intelligence offers insights on how we might design more effective thinking systems.

• Jorge’s Wager: Why AI Needs Information Architecture

  Why it’s still important to structure information in a world where AI exists.

• AI Needs Architects – That’s Why I’m Pivoting

  Intelligence demands structure, not hype. I'm stepping up.

• Book Notes: “Co-Intelligence”

  If you're only going to read one book about AI, this is the one.

• Horsey Horseless and the Challenge of AI-native Products

  What a weird 19th century product can teach us about designing for disruptive technologies.

• LLMs Give You Power User Abilities

  How to use AI to unlock deep capabilities latent in your tech stack.

• One Year of “Duly Noted”

  Updates and reflections on my book about note-taking, a year after its publication.

• The Resurgence of Plain Text in the Age of AI

  Reflections on the advantages of plain text and why it matters when working with LLMs.

• Beware the “Early Januaries” Content Pattern

  User-provided content and metadata can be unevenly distributed. Here’s what you can do about it.

• Book Notes: “Nexus”

  Overview of a book that explores the impact of information networks on human history.

• We’re in a Golden Age for Personal Automation

  Large language models are a boon for people who use automation frameworks.

• Scale Smart: AI-Powered Content Organization Strategies

  Two use cases of how large language models can help augment the work of information architects.

• Exploring the AI Solution Space

  Unlocking AI’s potential requires broadening our understanding of what’s possible.

• Seeing the Forest: Using Graph RAG for Information Architecture

  How to use knowledge graphs with retrieval augmented generation (RAG) to organize large content repositories.

• Envisioning (Near-)Future IA Design

  On the differences between using AI for automation vs. augmentation — and why information architects should explore both.

• Three Things I Know

  When living in times of radical technological change, look to things that have stood the test of time.

• The GREW Model

  A framework for leveraging the power of generative AI while keeping human oversight.

• Why I'm Excited About Generative AI

  Generative models aren't sentient entities but advanced symbol manipulation tools with vast potential.

• Empowering Information Architecture with AI: Making It Real

  An approach to using AI to re-categorize content in a 1,200-page website.

• Reconsidering the Role of AI: Valuing Process Over Output

  How to best use AI in both UX and personal knowledge management.

• Information Architects Must Embrace AI

  While generative AI has downsides, it also has great potential — especially for information architecture.

• The Impact of AI on Front-end Information Architecture

  Some thoughts about how AI might transform information architecture as we experience it in user interfaces.

• LLMapper: An AI Concept Mapping Tool

  Overview of my most recent experiment with AI: a tool for drawing concept maps from Wikipedia pages.

• Five Future Roles for Designers

  Possible career directions for designers working in an AI-driven world.

• Using AI as an Assistant to Organize Content

  How I used AI to categorize a year’s worth of podcast content.

• Empowering Information Architecture with AI: A Practical Experiment

  Overview of an experiment of using LLMs to categorize and tag over a thousand blog posts.

• Thinking With Words

  Writing is more important than ever — and it doesn’t just affect how you communicate, but also how you think.

• AI for Reading Notes

  One way LLMs like ChatGPT can help you take better notes about what you read.

• AI as Writing Partner

  LLMs can speed up the writing process. Here’s one way they can help.

• Artificial Intelligence for Information Architecture

  Looking beyond the AI hype to explore how LLMs might help us design better information architectures faster.

• Three Roles for Robots

  Examining different ways in which AI can augment your thinking.

• The Hair Metal Disclaimer

  New tools are changing how we create. How will you respond?

• TAOI: Metaphor Search

  An exciting new way to query web search engines.

• AI-driven “Design”?

  A thought on the current state of ML-driven design tools.

• Book Notes: “The Alignment Problem”

  An overview of Brian Christian's book about machine learning and values.

• Machine Intelligence and the Design of Complex Systems

  On the balance between top-down direction and bottom-up emergence in the design of complex systems.

• Causes of (and Remedies for) Bias in AI

  On the potential of AI to identify and amplify human biases.

• The Synthesizers In Charge

  On the challenge of integrating AI into our culture, with a historical perspective.

• Artificial Intransigence

  Video streaming recommendation algorithms sometimes fall short.

• The Treachery of Deepfakes

  It's becoming harder to distinguish between reality and its representations in information environments.

• The Next Twenty Years

  On the shift in focus from connecting people to enhancing their online activities — and the changes this will bring.

• Talking With the Information Machines

  Users can struggle as they adapt to novel semantic environments.

Podcasts

• How AI Is Built Podcast

  An interview with Nicolay Gerold on the relationship between information architecture and AI.

• The Informed Life with Karen McGrane

  Overview of a conversation about how generative AI might affect content management.

• The Informed Life with Chris Risdon

  A conversation with a master service designer about the role of technology in delivering cohesive experiences.

• The Informed Life with Jodi Forlizzi

  A conversation with a distinguished practitioner and academic about the changing role of design in light of AI.

• The Informed Life with Colin Eagan and Jeffrey MacIntyre

  Overview of my conversation with two consultants focused on designing personalized digital experiences.

• The Informed Life with Rachel Price

  A conversation about how information architects can help organizations use AI more responsibly.

• The Informed Life with Emily Campbell

  A conversation on why we need a pattern language for interactions with AI-powered systems.

• The Informed Life with Michael Anton Dila

  Overview of a conversation about a new initiative that aims to redefine AI governance through inclusivity.

• The Informed Life Year in Review: 2023

  A compilation of themes that emerged in my podcast in 2023, partly curated by AI.

• The Informed Life with Alice Albrecht

  An overview of my conversation with the founder and CEO of re:collect, a new AI-driven tool for thought.

• The Informed Life with Lorenzo Bernaschina

  A summary of my conversation with the developer of Gems Notes, an AI-driven smart note-taking tool.

• The Informed Life With Carol Smith

  An with researcher Carol Smith on the benefits, limitations, and potential dangers of AI technologies.