Chapter 13 - The Materiality Dial: Calibrating Effort and Autonomy
Loop Engineering by Fabio Correa, the Applied AI foundation volume on working with AI through framing, feedback, verification, memory, and human judgment.
Last updated 2026-10-25. This page collects the two sources Chapter 13 cites, Runkle and Kahneman, and links into the extended materiality patterns by industry domain.
The question this chapter takes up
How much loop does a piece of work earn, and what decides: the stakes, or the person accountable?
Put it to work
Use the materiality patterns to choose effort and autonomy for one real task before starting it.
Where practice stands
The chapter's October 2026 snapshot: auto-approve options are spreading across agent tooling, and approval surfaces are outrunning reviewer attention. The gating principle, calibrate authority to stakes, was already working practice by April 2026, and a standing bounded grant, informal and tracked in a runbook, existed in practice by September 2026. Chapter 16 presents the telemetry behind the pattern, Anthropic's approval data [D85]; see the Chapter 16 page, where that source is linked.
Key sources
Chapter 13 cites in the print book: D23, D46. Entries below are the sources this page draws on; some are related literature cited elsewhere in the book rather than in this chapter.
- D23 - Runkle, S. (2026, June 16). The Art of Loop Engineering. LangChain. langchain.com/blog/the-art-of-loop-engineering
- D46 - Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux. ISBN: 978-0374275631
Primary-source links
- Runkle (2026), "The Art of Loop Engineering" - the chapter's epigraph source [D23], a practitioner statement of loop design from LangChain.
Related resources
The materiality patterns by domain apply the 3×3 dial across software engineering, healthcare, legal, policy, education, customer support, and financial operations. Each domain entry shows where the dial sits and which compensating controls the domain demands.