This university helps you start working in each world, from human to robot. Learning here runs top-down — begin on the rooftop and work your way to the basement. Click any floor for hand-picked videos, channels and websites.
Start with how you think and work.
From Tool to Collaborator
The way we think about AI is shifting fundamentally. Understanding this shift is the foundation for everything else.
A concept articulated by Jeremy Utley (Stanford d.school): AI functions best not as an answer machine but as a thinking companion. A strategic thought partner engages with you across every phase of a project—planning, research, discussion, creation, and feedback—helping you think better rather than thinking for you.
Cyborgs & Centaurs
Researchers at Harvard and MIT tracked 244 consultants working with AI on a real business task and found three distinct patterns of use. How you work with AI shapes both the quality of your output and what you learn along the way.
Self-Automator — "Abdicated Co-Creation"
Self-automators hand the task over to AI almost entirely, delegating the analytical and evaluative thinking along with the busywork. The result comes fast and looks polished, but it lacks depth. In the study, self-automators (27% of consultants) produced the least persuasive work and showed no skill gains — a warning for trainees who let AI do the thinking for them.
Centaur — "Directed Co-Creation"
Like the mythical half-human, half-horse, centaurs keep a clear dividing line between human and machine work. They know what questions they want answered, ask targeted questions, and stay guided by their own expertise. Centaurs (14%) had the highest accuracy of the three groups — and deepened their own domain expertise while working.
Cyborg — "Fused Co-Creation"
Cyborgs weave AI into every stage of the work — a continuous conversational back-and-forth, probing suggestions, taking some advice and pushing back on the rest. Cyborgs (60%) matched centaurs on persuasiveness. They gained little domain expertise but built a different valuable skill: knowing how to solve problems with AI.
The takeaway for medical educators
- Avoid self-automating tasks that are core to your role or high-stakes — that's where depth, accuracy, and learning are lost.
- Work as a centaur when accuracy matters and you want to build your own expertise; work as a cyborg when you want AI deeply in the loop and are building AI-collaboration skill.
- Sources: HBS Working Paper 26-036 (Randazzo, Lifshitz-Assaf, Kellogg, Dell'Acqua, Mollick, Candelon & Lakhani) · MIT Sloan: 3 Ways to Use AI
Co-Thinking
Co-thinking is interactive collaboration—you and AI working together through conversation. The three levels below (Capture, Challenge, Coach) represent increasing depth of collaboration, from you leading the process to AI guiding it.
At this level, you're firmly in control. You have ideas, tasks, or information—AI helps you capture, organize, and transform them. This is where most people start, and it's immediately useful.
1. Voice Dictation
Speak your thoughts naturally and let AI transcribe, organize, and respond.
Why it matters: When you speak instead of type, ideas flow more freely. Typing naturally triggers self-editing—you stop to fix phrasing, second-guess word choices, and lose momentum. Dictating bypasses that filter, letting your thoughts come out in a raw, uninterrupted stream that AI can then help you organize and refine.
2. Long running Thought Partner
Use the same the chatbot thread over days and weeks storing your thoughts.
Also consider as a “Second Brain” with a persistent knowledge repository that AI can reference across conversations. You add documents, past decisions, policies, and context—so can AI help you make connections and retrieve relevant information in your thought process.
A second brain is an external system that stores your knowledge, ideas, and reference materials in a way that's searchable and connectable. With AI, this becomes interactive—you can ask questions of your accumulated knowledge, not just search it. The term was popularized by Tiago Forte.
At this level, AI becomes an active thinking partner. Instead of just capturing and transforming your ideas, it challenges them, offers alternatives, and helps you see blind spots.
1. Non-Obvious Solutions
Explicitly ask AI to go beyond the obvious and explore less traveled paths.
When you present a problem, AI's first suggestion is often conventional. Push it to explore alternatives you haven't considered.
Why this works: AI has absorbed countless approaches to common problems. By asking it to distinguish obvious from non-obvious, you filter for insights that are actually new to you.
2. Sparring Partner
Ask AI to argue against your position to stress-test your thinking.
This isn't about AI being right—it's about stress-testing your thinking before you commit to a decision or communicate it to others.
At this advanced level, you flip the dynamic: instead of you driving the conversation, AI guides you through a structured process. This is powerful when you're stuck, unclear on what you need, or facing a complex problem.
1. Planning Mode
Let AI structure a project before you dive into execution.
Describe what you want to accomplish; AI creates a plan, timeline, and identifies potential obstacles.
The power of "don't start yet": By explicitly asking AI to plan before executing, you get a roadmap you can review and adjust rather than outputs you have to redo.
2. Iterative Research
Have AI self-evaluate its output and improve through testing and iteration.
Instead of accepting AI's first output, create an iterative improvement cycle where AI reviews its own work, identifies gaps, and refines the result. This mirrors the scientific method: generate, test, improve, repeat.
How this works: After AI produces initial research or analysis, ask it to read what it created, identify what questions or concerns the output raises, gather additional information to address those gaps, and then incorporate the new insights into an improved version.
Why this matters: AI's first attempt is often surface-level. By building in self-evaluation and iteration, you get deeper, more nuanced outputs that anticipate follow-up questions and address weak points before you even ask.
Rebuild workflows as AI-centric
Adding AI into existing workflows is just the first step. The hard future work is redesigning your specific workflows from scratch with AI at the center. Using AI as thinking partner is a start.
Try Co-Creating
Co-Creating
Context is Everything
Context is the difference between thoughtful and lazy AI use. When you get a generic, sloppy AI response—what went wrong? Usually, you expected AI to read your mind. It can't. But give it proper context, and everything changes.
Key Concepts: Tokens, Context & Collaboration
The Big Shift: Modern AI can now handle very large amounts of information (called "tokens"). This means you can give it extensive context—entire documents, multiple files, detailed background—and it will use all of it to give you better responses.
What Are Tokens?
Tokens are the units AI uses to process text—roughly equivalent to words or word pieces. Early AI could only handle a few thousand tokens (a few pages). Modern AI can handle 100,000+ tokens (entire textbooks worth of content). This means you can now upload entire rotation evaluation forms, curriculum documents, or detailed case histories and AI will use all of it.
The Mind-Reading Problem
When AI gives you bad output, ask yourself: "Did I expect AI to know things I never told it?" Generic prompts produce generic results. Specific context produces specific, useful results. AI can't know your department culture, this resident's history, or your communication style unless you tell it.
From Oracle to Collaborator
Stop thinking of AI as a magic oracle that should just "know" what you need. Instead, treat it as an informed collaborator—like briefing a new colleague. The more you share about your situation, goals, and constraints, the better it can help.
Three Steps to Thoughtful AI Use
Here's a systematic approach to working with AI that consistently produces better results.
Future Co-Working
Co-Working
Co-working is what happens when AI moves beyond back-and-forth conversation to independently completing tasks for you — like a resident you can delegate - and you mostly focus on giving direction and feedback.
This already exists today: Tools like Claude Cowork and Google Antigravity let you hand off a task to AI — it works on it independently while you see patients, answer emails, or do anything else.
"Vibe coding" means describing what you want rather than figuring out how to make it. You tell AI the end result you need; it figures out the steps. This goes well beyond code — you can describe what you want and get slide decks, research summaries, data analyses, or any complex output without knowing how to build it yourself.
Manage Agents and People Together
The future of AI work isn't a chat window you stare at. It's a project board you manage — with AI Agents and people working side by side.
You manage a project board where AI Agents and people both own tasks, asking each other for approval and help as the work moves forward. You tell agents what you need done and how to know if they got it right — not how to do it. They work independently, hand off to teammates, and come back when they're finished or need input.
One Board. Agents and People.
Imagine this: You open your project board in the morning. An AI Agent has finished a research summary on new compliance requirements and tagged your program coordinator to fact-check it. Another drafted a schedule optimization and is waiting on your approval before notifying residents. A colleague assigned an agent to prepare talking points for your faculty meeting and flagged two for your review. Agents and people ask each other for help, hand work back and forth, and escalate to you only when your judgment is needed. You review, approve, redirect — and assign what's next.
Today
You chat back and forth with AI — your full attention is needed
Emerging
You assign a task to AI — check back when it's done
Future
AI Agents and people share one project board — assigning tasks, requesting approvals, and helping each other
Less doing. More directing
We are moving from sitting with AI while it finishes a task, to assigning work to teams of AI Agents and checking in to set direction and judge quality. Your role shifts from doing the work to deciding if the work is good.
Interactive Lecture
A lecture in three phases — connect first, confront and reframe in the middle, and consolidate together at the end.
Start of Session — Connect with Your Audience
Open by meeting learners where they are. Tap into common experiences, address their fears directly, assess what they already know — and let them voice it. Connection before content.
Main Session — Confront and Reframe
Confront what learners don't know, unpack their misconceptions, and give them new frameworks to think with. Keep it interactive: use check-ins every 10–15 minutes to surface questions and keep everyone engaged.
End of Session — Review, Reflect, Future Study
Close by reviewing the key points, reflecting on what shifted, and pointing toward future study. Do some of this together — shared reflection cements the learning.
What matters to you
Why should you specifically care about this?
What are the AI advances creating recent headlines?
What AI means for our future work, identity, society?
What are responsible use and needed guardrails?