↓ Skip to main content
  1. The Lexicon/

Training half-life

Ben Schmidt, PhD
Author
Software architect and founder. I build with AI agents every day and write about the hard part, which is knowing what to ask, and catching the machine when it’s confidently wrong.
Table of Contents

Training half-life is how long it takes for half of what someone learned in training to leak back out, with no reinforcement.

It is shorter than anyone selling you a workshop wants to admit. The forgetting curve is real and it is steep. One-and-done training has a brutally short half-life; spaced, low-dose practice is the only thing that extends it. The slide deck is not the intervention. The reps after it are.


Related Reading

#

The AI & Learning Field Report

Field notes on AI and how people actually learn.

What is real, what is hype, and what the evidence actually says. The reports land here first; subscribers get them in their inbox. No spam, no funnel theater.