Judgment

I had the pleasure of corresponding via email with Andy Nian, cofounder and head of growth at PracHub, a site that catalogs technical interview questions reported by candidates. He wrote to me in August asking whether I would cite his AI-assisted question set in a piece I had written about why interviews should measure judgment rather than output. I came back asking for his data instead, and he graciously answered a few questions on the record: figures, clear caveats about what those figures could not support, and three public examples. Almost no founder does that. One of his answers has stayed with me since, and most of what follows is an argument with it, in the good sense. ...

I recently came across a brilliant comedy clip by Deepak Chandran about how we use AI to write emails. Someone has a simple message. They give it to AI, which turns it into a beautifully structured, professionally worded, eighteen-page document. The recipient can’t be bothered to read it, so they ask AI to summarize it. We’ve built a remarkably sophisticated machine for turning one sentence into eighteen pages and back into one sentence. ...

Reading the “Man and Machine” chapter of Zero to One, I was struck by how confidently Peter Thiel and Blake Masters argued that computers would complement humans rather than replace them. Their reasoning was straightforward: computers excel at processing data; humans bring judgment. Build businesses that combine those strengths. Then AI advanced. In a December 2025 interview with The Spectator, Thiel described AI as “probably more of a substitute than a complement” to human labour. ...

During a train ride, I built a website for our family vacation. I used ChatGPT on my phone. After about an hour, I had a site with a route for each day, photos, maps, riddles for the kids, and ideas for things to do. The users were six people: my family. I had no business plan. I did not want to start a travel company or sell subscriptions. I did not want to make it useful for thousands of other families. I wanted something that fit our trip. ...

There is a pattern in senior resumes and LinkedIn profiles that makes me more uncomfortable every year. Someone “grew revenue by 38%.” Someone else “reduced costs by $4.2M.” Another “increased engineering productivity by 60%.” Sometimes I know the work behind these lines well enough to know that the link between the person and the number was much weaker than the sentence suggests. It is tempting to call this a problem of dishonest people. My view is that it is worse than that. I believe we built a market that rewards the performance and punishes the truth. In my experience, the candidate who writes “grew revenue by 38%” is more likely to get the interview than the one who honestly describes a shared effort. When the reward for exaggeration is high and the chance of being checked is low, I do not see exaggeration as a personal failure. I see it as the rational response to the system we built. ...

Dario Amodei, CEO of Anthropic, has published an unusually direct argument about the future of AI development: the companies building the most capable models need to slow down. He is not calling for them to stop, or to give up the benefits AI may bring. He is asking them to leave enough time to understand, evaluate, and control the systems they are making more powerful. What makes this worth reading is who is saying it. Amodei runs one of the companies competing to build those models. He faces the same pressure to move faster as everyone else, and he is arguing that the pace itself has become part of the problem. ...

Anthropic published a new piece last week, The AI-Native SDLC Playbook. It is long, detailed, and worth the time. It also lines up closely with what I have been writing here for the better part of a year. The argument at its center takes one sentence: code is no longer the bottleneck. Everything in the document follows from it. Once agents write most of the implementation, the constraint moves to the stages on either side, the ones still running at human speed. Approval queues build. Controls sized for human output stop matching what arrives at them. Governance that meets monthly cannot govern work that ships hourly. ...

On a recent episode of Aakash Gupta’s podcast, Oji Udezue ran a product idea through a viability gate he had built as a Claude Code skill. The idea was a tool called Standup Zero that reads Slack comments and assembles a daily standup digest. The gate scored it on six dimensions and returned three moderates. Problem clarity and urgency, moderate, because workflow convenience is not a deep problem. Target user definition, not strong. Differentiation, not strong. Competitive landscape, strong, which on this scorecard counts against you, because it means everyone can build what you are building. ...

A June 2026 NBER working paper matched more than 100,000 GitHub developers against their AI usage telemetry. Developers using autonomous agents committed 180 percent more code. Their releases rose 30 percent. The gain is enormous where code gets written, and it decays at every step after that. Marty Cagan named the phenomenon last week: the AI productivity paradox. He also points out that while nearly everyone now agrees it exists, almost nobody agrees on why. ...

I recently needed to understand a long, cross-departmental process. It was documented across a huge slide deck, a few forms, and half a dozen fields scattered across internal systems. No single source of truth. In my experience, the right tool for this is a checklist. Not a dashboard, not an assistant. Just a simple list of boxes to check in the right order. So I used Claude to read through the materials and produce a tight summary of the process and its steps. I reviewed it, then asked Claude to turn that summary into a checklist draft. I refined it. I’m using it now, and I’ll keep refining it as I go. ...