Skip to content
Module
Module 5 of 5Lesson 1 of 2~18 min

The "delegate / verify / keep" grid

Handing everything to AI exposes you to costly mistakes, while handing it nothing wastes valuable time. This lesson gives you a decision grid to sort each task in your job into delegate, verify or keep, and to plan the right check for what you hand over.

Take this lesson: your first course is freeSee the full syllabus

Lesson objective

By the end of this lesson, you will be able to sort a task into delegate, verify or keep using four questions (cost of an error, verifiability, model property involved, value of doing it yourself), and to name the planned check for each task delegated with control.

Topics covered

  • delegating to AI
  • AI decision grid
  • responsible AI use
  • AI productivity

Where it fits

Deciding what to delegate

For each of my tasks, what do I hand to AI, what do I verify and what do I keep?

Lessons in this module

  1. The "delegate / verify / keep" grid (this lesson)
  2. Workshop: your grid on ten tasks, the exit kit and the application plan

What you will learn in this module

This lesson is part of the course Understand what LLMs do well, and where they fail

  • Anticipate hallucinations, sycophancy, poorly calibrated confidence and run-to-run variability, and choose the right check for each case.
  • Locate what the model knows (training data, knowledge cutoff, uneven coverage) and what web search, provided documents and connectors change.
  • Explain the context window as working memory (limit, degradation over long contexts, no memory across conversations without a dedicated feature) and derive working rules from it.
  • Sort the tasks of your job into "delegate / verify / keep" based on the cost of an error, how easy it is to verify, the model property involved and the value of doing it yourself.

Related courses