Every Product You Ship Has an Anthropology with Taylor Black from the AI and Venture Ecosystems at Microsoft & the Leonum Institute for AI and Emerging Technologies

In this episode, I sit down with Taylor Black, who leads AI and venture ecosystem work in Microsoft’s Office of the CTO and brings a deep “better-formed humans” conviction to the AI conversation. We draw a hard line between intelligence (statistical knowing) and insight—where a human knower makes the categorical leap to “this is true,” what Taylor calls grasping the “virtually unconditioned.” We talk about how LLMs can pattern experience to help us reach insight but can’t replace judgment, and why venture building should be anthropology-first: every product ships an implicit view of the human person. We cover the risk of using AI as an oracle that shortcuts productive struggle (especially in education), the importance of humility and sustaining the tension of inquiry for product-market fit, and Taylor’s “cold storage” idea—using AI as a context engine to shelve, monitor, and revive ventures when conditions change. We close on the key builder question: not what can we build, but what ought we build?

02:08 Intelligence vs Insight

02:58 Statistical Knowing

05:02 Virtually Unconditioned

09:02 Using AI Without Outsourcing Judgment

11:39 Formation in Venture Building

14:19 Anthropology First Product Design

17:56 Hard Rules for Human Flourishing

18:53 What Ought We Build

21:20 Humility and Idea Flow Playbooks

24:19 Probabilistic Products and Implicit Anthropology

28:48 AI as Oracle in Education

34:29 Cold Storage for Venture Ideas