In over thirty years in business, two technological shifts have fundamentally changed my work. The first was the web in the mid 1990s, which led me to leave architecture and focus on what we now call UX. The second is happening now – and it may be bigger. AI is poised to reshape how we interact with information, much as the web did – perhaps more so. This isn’t just another technological shift. It’s a fundamental change in how people understand their world and how businesses create value. But without structure, AI risks creating more confusion than clarity. That’s why I’m pivoting my consulting practice to what I’m calling Architecture for Intelligence. The phrase is intentionally ambiguous. Is the target human intelligence or artificial intelligence? The answer is “yes.” Thoughtfully structured information amplifies intelligence, human or artificial. Done right, it creates a virtuous cycle: smarter humans design smarter systems, which make humans smarter. But it won’t happen out of the box. I’ve worked enough with current AI systems to understand their capabilities and constraints. They’re astonishing in many ways, but they struggle to create stable, cohesive information environments – which people need to make skillful decisions. At least not without help. And that’s why I’m so excited: Because with a bit of structure, AI-powered systems can unlock previously unimaginable abilities — including the ability to produce smarter information environments. This sounds a bit academic, but the implications are huge. Our expectations about what constitutes good service will change. The speed, scale, and degree of personalization will be different. Businesses will need to rethink their strategies as the market shifts to the new paradigm. Those who do it well stand to benefit like never before. But individual humans must benefit too. This shift should bring value to people first; creating value for customers will create value for businesses. And here again is why we need Architecture for Intelligence. Architects don’t design structures for their own sake; we do it to serve human needs. This simple idea risks being lost in the mad rush to implement AI-powered systems. But AI doesn’t reduce complexity out of the box. It doesn’t generate clear reference frames for people to get their bearings. And done poorly, it can significantly degrade trust – or worse. AI doesn’t change the fact that clear, usable, and strategically aligned products require architecture. If anything, it increases the need for carefully considered structures. The shift goes both ways: AI needs information architecture in order to be usable, understandable, and trustworthy. IA itself is being transformed by AI. My new focus embraces both angles by 1) helping clients design more intelligent systems and 2) developing tools and methods to architect information more intelligently. All in service to amplifying human intelligence – from somebody with firsthand experience with the technology and over three decades of perspective. Specifically, I’m offering three kinds of engagements: Strategic consulting: Advising leaders on how to structure information to support AI-driven systems that drive real value. Product & web IA: Helping internal design and product teams deliver AI-driven products and websites that are easier to navigate, use, and maintain. Training: Showing teams how to use IA principles to design better smarter products and make better strategic decisions. I’m also working on AI-powered tools to do information architecture more effectively. And of course, you can expect the bulk of my writings and presentations to focus on AI from now on. (As they’ve done for much of the last year.) Which is to say, like AI itself, this change isn’t coming: it’s already here. Organizations that architect intelligence are poised to benefit like never before. And by centering human needs, you can ensure the benefits are felt widely. But you must start now. I’m more excited to be in tech now than at any time in the last thirty years. AI’s potential – for good and bad – is vast. Let’s make it good. If this resonates, please get in touch.

Week 7 of my humanities crash course had me exploring ancient Mesopotamia with a side trip to northern India. I also watched an Iranian film that had me pondering the meaning of life. Readings This week, I read two short ancient texts: the Epic of Gilgamesh and the Dhammapada. Let’s tackle them in order. I’d never read any ancient Mesopotamian literature, so this was all new to me: the pantheon, story, style, etc. were thrillingly unfamiliar. Gilgamesh is around 1,500 years older than Homer, and it shows: there are lots of repetitive passages and what felt like archaic writing. But human nature hasn’t changed much in 4,700 years. People still love, hate, drink, eat, etc. – and they still fear death. Gilgamesh is the awe-inspiring, despotic king of Uruk. The gods answer his beleaguered subjects’ prayers in the form of Enkidu, a rival who becomes Gilgamesh’s friend. They embark on several heroic exploits and end up pissing off the gods. As a result, they condemn Enkidu to death. Despondent and fearing for his own death, Gilgamesh goes in search of the secret of immortality. His travels take him to Utnapishtim, immortal survivor of the great flood. Our hero finds a plant that restores youth, but loses it. By the end of the story, he accepts his fate as a mortal. The story moves fast and is surprisingly engaging. It includes early versions of ideas that would resurface later in the Bible. (Most obviously, Noah and the flood.) There’s also some material that probably wouldn’t pass muster in our prudish time. The Dhammapada is one of the central Buddhist scriptures. I was familiar with several of these texts but hadn’t read the whole thing. Gioa notes that he recommended it because of its length, but there are also obvious connections with Gilgamesh. For example, several verses in the Dhammapada deal with attachment. For example, here’s verse 215: From affection comes grief; Gilgamesh suffers from such an attachment. Here’s the moment of Enkidu’s death: He touched his heart but it did not beat, nor did he lift his eyes again. When Gilgamesh touched his heart it did not beat. So Gilgamesh laid a veil, as one veils the bride, over his friend. He began to rage like a lion, like a lioness robbed of her whelps. This way and that he paced round the bed, he tore out his hair and strewed it around. He dragged off his splendid robes and flung them down as though they were abominations. He wishes to hold on: Then Gilgamesh issued a proclamation through the land, he summoned them all, the coppersmiths, the goldsmiths, the stone-workers, and commanded them, ‘Make a statue of my friend.’ The statue was fashioned with a great weight of lapis lazuli for the breast and of gold for the body. A table of hard-wood was set out, and on it a bowl of carnelian filled with honey, and a bowl of lapis lazuli filled with butter. These he exposed and offered to the Sun; and weeping he went away. You can probably relate if you’ve ever lost someone dear. Human nature. Audiovisual Gioia recommended Stravinsky’s Rite of Spring and Wagner’s Overtures. I’m very familiar with both so I didn’t spend much time with either this week. For a new take on one of these familiar classics, check out Fazil Say’s astonishing piano version of the Rite of Spring. Here’s a short portion: Gioia also recommended looking at ancient Mesopotamian art. I didn’t spend as much time on this as I would’ve liked. That said, this introductory lecture provided context while highlighting major works of art and architecture: I took a different approach to cinema this week. Rather than go by an AI recommendation, I went down the old-fashioned route. (I.e., Google.) Specifically, I thought this would be a good opportunity to check out Iranian cinema. I’ve heard good things about Iranian films, but had never seen one. Googling led me to this article on Vulture. After reading through the list, I picked Abbas Kiarostami’s TASTE OF CHERRY. Yet again, I’ve gravitated towards a film about a middle-aged man in despair. (Is the Universe trying to tell me something?) Kiarostami effectively uses a minimalist style to explore what makes life meaningful despite (or perhaps because) of its finitude. Reflections There’s a pattern here. This week’s works dealt with core issues people have grappled with since we became people. The big one: how do we deal with death? Not just the impending death of everyone we love, but our own. Gilgamesh offers the traditional “Western” answer: “I can’t even.” So, fight it! He looks for a MacGuffin that’ll let him go on living and perhaps brings his loved ones back. It’s an idea that has had many progeny in our mythologies. And it’s not just the stuff of fiction: the impulse is still alive and well. (Pardon the pun.) The Buddha offers a different approach: non-attachment. It’ll be easier to let go if you don’t become enmeshed with things, people, and your own sense of being. So you train your mind so it won’t hang on. (Even the idea of “your mind” is suspect.) The price: not feeling either extreme. No despair, no elation. Kiarostami’s film suggests a third approach: accepting the inevitability of death while reveling in the experience of being alive. (You could argue this is part of the Buddhist way as well.) I won’t say more in case you haven’t seen TASTE OF CHERRY, but suffice it to say the film employs a clever structural trick to wake you from your slumber. Grappling with these kinds of issues is the point of studying the humanities. Yes, I know you’re busy. I’m busy too. But some day, the busyness will stop – as will everything else. I’m committed to living an examined life, and that requires thinking about stuff we’d rather put aside so we can get on with the next Zoom meeting. Notes on Note-taking I’m also committed to the other point of this humanities project: learning how to learn better in this AI age. This week, I continued tweaking my note-taking approach. I took notes in the Drafts app as I read, building an outline as I go through the week. I wrote down the main points I learned and things I’d like to share with you. I then elaborated this outline on one of my morning walks. My mind works better when my body is moving and clear from the day’s detritus. I also tweaked my note taking approach around the readings. I had an LLM summarize the reading and then used that as a refresher to write a summary in my own words. I’ve done the same in previous weeks. What’s different now is that I then pasted my summary into a ChatGPT window with a simple prompt: I read The Epic of Gilgamesh. What is wrong with this description of the story?: Gilgamesh is king of Uruk. He’s described as the strongest and most beautiful man in the world. He’s also something of a despot. He befriends Enkidu, a wild man who is almost as strong as Gilgamesh. They go on several adventures, which entail opposing the wishes of one of the Mesopotamian deities. Eventually, the gods are angered and decree Enkidu must die. Grief-stricken, Gilgamesh goes in search for the secret of eternal life, only to learn that human lives are limited. He returns to lead his people with this newfound wisdom. The LLM offered a helpful response that clarified nuances I’d missed: Gilgamesh wasn’t “something of a despot”; he was a tyrant. The gods created Enkidu as a counterbalance to answer his subjects’ calls for relief. The taming and channeling of this force of nature through initiation into human pleasures is an important aspect of the story I’d left out. Details about Gilgamesh and Enkidu’s transgressions against the gods. (These seemed less relevant for a high-level summary.) The fact Gilgamesh isn’t just searching for immortality because he’s grief-stricken over Enkidu’s death; he’s also fearing for his own life. The end of my summary was wrong; the book doesn’t suggest Gilgamesh changed as a result of his experiences. This last point is important. In writing my summary, I made stuff up that wasn’t in the book. I attribute my error to the fact I expect closure from my stories. Gilgamesh precedes Aristotle’s Poetics; its authors were under no compulsion to offer the hero a redemption arc. Which is to say, humans hallucinate too – and LLMs can correct us. Up Next Next week, we’re reading ancient Egyptian literature. I couldn’t find an ebook of the text suggested by Gioia, so I’m going with another Penguin book, Writings from Ancient Egypt. I studied some Egyptian architecture in college, and look forward to revisiting this part of the world and its history. Check out Gioia’s post for the full syllabus. I’ve started a YouTube playlist to bookmark all the videos I’m sharing in this course. And as a reminder, I’m also sharing these posts via Substack if you’d like to subscribe and comment.

Co-Intelligence: Living and Working with AI by Ethan Mollick Portfolio, 2024 Over the weekend, I caught up with a friend I hadn’t seen in years. When I explained that most of my work these days is focused on AI, he asked if there was one book he could read to understand how to better use AI. This is the book I recommended. Why? Because it offers a clear explanation of how the technology works, thoughtful explorations of what it means for us, and practical suggestions for using it to help with common tasks. AI isn’t like other technologies. As Mollick puts it, We have invented technologies, from axes to helicopters, that boost our physical capabilities; and others, like spreadsheets, that automate complex tasks; but we have never built a generally applicable technology that can boost our intelligence. The book is divided into two parts. The first explains the fundamentals, including how transformers work and the challenges inherent in aligning AI goals with human goals. Mollick lays out four principles for working effectively with AI: Always invite it to the table: AI is a general-purpose technology; it behooves you to try it for different ends. Be the human in the loop: assume AI will need supervision and guidance; provide it. Treat AI like a person, but tell it what kind of person it is: don’t fall into the trap of anthropomorphizing AI – but it’s helpful to suspend your disbelief. Assume this is the worst AI you’ll ever use: the technology is improving fast; assume the things that are difficult or impossible today will be doable in the future. The second part explains how to use AI effectively by seeing it through the lens of five possible roles: As person: even though we should remember AIs aren’t people, it can be useful to ask it to simulate human attributes such as personality and particular expertise. As creative: AI can assist us in our creative efforts or replace our creative functions outright. You’ll be better served by using AI for creative augmentation rather than replacement. As coworker: AIs can handle information tasks at much greater speed and scale than people. Knowing what they can and can’t do well can help you get the most out of them in a work context. As tutor: AIs can offer personalized instruction and draw from a larger corpus than what any human can. That said, they can also hallucinate. As coach: AIs can help us find patterns in data and help nudge us towards accomplishing our goals. In all five cases, the emphasis is on collaborating with AI as a partner rather than a replacement. The book offers concrete examples of prompts and the subsequent responses. Some are about the writing of the book itself, a nice meta-touch. The last chapter speculates about four possible futures, ranging from a scenario where the technology stalls to the opposite extreme where AI becomes all-powerful. The middle scenarios – slow growth and exponential growth – seem likeliest to me. As I implied above, this is the best primer I’ve found on what most people currently understand by AI. The book is clear about the risks and challenges inherent in the technology, but also practical, informative, and philosophical. Much of its advice validates my experiences with the technology. Co-Intelligence: Living and Working With AI

In episode 3 of the Traction Heroes podcast, Harry and I discussed the relationship between decision-making and data. Is it ok for some decisions to be made intuitively? What role do emotions play in decision-making? How do you deal with stakeholders who demand that decisions be backed with data? Tune in to find out!