Research Insights #40 materials: Anthropic—what remains for people in a company of agents (#Management)

The materials for solo Research Insights Made Simple episode #40 from October 19, 2026 are ready. I explored “How employment is changing at Anthropic”, a study of work inside Anthropic: agents produce more and more, while people still need to decide what to accept, what to learn, and where to go next.

In the episode I covered:

🔸 What the measurements show The authors compare AI use, hiring, and work outcomes. These are observations within a rapidly growing company: they do not prove that agents caused every change, and code volume alone does not measure value.

🔸 Where human attention goes By the authors’ calculation, reading all agent output would not fit into an engineer’s working week. People respond with instructions, checks, and monitoring tools. Even former individual contributors now have to manage agents’ work.

🔸 Where newcomers can gain experience If agents have already taken the simple tasks, where will people acquire the understanding needed to review complex decisions? The episode offers no ready answer; learning needs to be designed alongside automation.

🔸 What employees get I discuss moves into new roles, time for learning, and sharing in the gains. The opportunities at a growing Anthropic cannot automatically be extrapolated to a company where there simply is no next role for an employee.

I particularly recommend this to anyone whose discussion of AI still ends with the amount of generated code. Here, that is only the beginning.

Episode materials: 📌 Episode page with timestamps and episode slides 🎬 Video: YouTube and VK Video 🎧 Audio: Podster and Apple Podcasts 📖 Text summary and my annotated PDF (where I recorded my thoughts in the margins while reading)

How are learning and people’s responsibilities changing in your organization as some work moves to agents?

#ResearchInsights #AI #Agents #Management #Engineering