I recently finished teaching AI and Strategy, a condensed one-week elective for the Executive MBA program at the University of Michigan Stephen M. Ross School of Business.
It was exciting to teach because the field is moving so quickly. The course already felt different from the version I taught in November. Then, generative AI was central to strategy. This time, agents—especially coding agents—had moved to the center of the conversation. They change what firms can build, what work can be delegated, what managers need to understand technically, and where accountability gets tricky.
The course moved from micro to macro: how AI changes strategic decision-making; how firms should respond when AI reshapes competition and advantage; and what more capable systems mean for work, organizations, regulation, safety, trust, and society.
There was too much to cover. I could easily imagine a much longer course. That is a good problem to have.
The students brought real questions from their work: where AI puts pressure on existing business models; which tasks to automate or augment; which data bottlenecks matter; when an AI product is defensible rather than just a wrapper; who should own AI inside the firm; and what leaders should do when the technology is useful but still unreliable.
Those are the questions that make this topic worth teaching. AI strategy is not “How do we use ChatGPT?” It is: What decisions change? Who captures the value? What becomes cheap? What becomes scarce? What should the firm own? And what should humans remain accountable for?
Many thanks to Michael Olenick of vstrat.ai, Sameer Soleja of Molecule Software, and Akshay Kapoor of McKinsey for joining as guest speakers. Each conversation was deep, practical, and grounded in what executives are actually facing.
I’m sharing the syllabus in case it is useful to others teaching, studying, or leading through this moment.
—