The sources

Chapter 17. Courage

3 sources behind this chapter. Each one carries a note on what it is doing in the argument.

The chapter itself, for readers of the book

1. Research: Why You Shouldn’t Treat AI Agents Like Employees

   Kropp, Matthew, Julie Bedard, Emma Wiles, Megan Hsu, and Lisa Krayer Harvard Business Review, 6 May 2026

   The courage chapter uses this experiment to show the cost of pretending an AI agent is an employee. Managers reviewed less carefully and felt less accountable, without gaining better adoption.

   Open this source Read the original

   Kropp, Matthew, Julie Bedard, Emma Wiles, Megan Hsu, and Lisa Krayer. ‘Research: Why You Shouldn’t Treat AI Agents Like Employees.’ Harvard Business Review, 6 May 2026.

2. Bot versus human traffic data

   Cloudflare Radar; Matthew Prince Cloudflare Radar / X, 3 June 2026

   Cloudflare Radar’s bot-versus-human HTTP-request data. CEO Matthew Prince announced the crossover on 3 June 2026; the contemporaneous snapshot was approximately 57.5 per cent automated requests to 42.5 per cent human requests.

   Open this source Read the original

   Cloudflare Radar. ‘Bot versus human traffic data.’ June 2026; Prince, Matthew. X post announcing the bot/human crossover, 3 June 2026.

All the sources, chapter by chapter