Tech Lead Journal

From Technical Debt to Triple Debt: The Hidden Cost of AI-Generated Code

52 snips
Jul 27, 2026
Margaret-Anne Storey, a professor and researcher on developer experience and generative AI, outlines her “triple debt” model. She discusses technical, cognitive, and intent debt and how AI accelerates loss of shared understanding. She describes cognitive surrender, developer fatigue from agent workflows, and practical fixes like Cognitive Tours and strategic friction.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Cognitive Debt Erodes Team Memory

  • Cognitive debt accelerates transactive memory loss: teams forget who knows which parts of the system.
  • This loss compounds when many agents produce code in parallel and no one documents ownership or rationale.
INSIGHT

AI Can Fix And Fuel Cognitive Debt

  • AI can both repair and create cognitive debt: it helps explain code but also enables faster accumulation of unexamined changes.
  • Repair works only if teams pause, read code, and use AI as a comprehension aid rather than full automation.
ADVICE

Write Down Intent Not Just Code

  • Record intent explicitly; you cannot always recover why a decision was made just from code.
  • Write down decisions, what you decided not to do, and rationale so agents and future humans have the needed context.
Get the Snipd Podcast app to discover more snips from this episode
Get the app