Framework

Chain-of-thought prompting

The question it answers

When a manager asks AI to work through a multi-step business problem, how should they prompt it to reason step by step so each step can be checked before trusting the answer?

Origin
Wei et al., 2022
Diagram type
Process by phase
Level of analysis
Individual
Topics
AI for leaders
Glass sphere with a glowing neural-network core, icon for AI for leaders

Diagram

Chain-of-thought prompting

  1. Three steps
  2. State the problem with all the facts
  3. Ask the model to reason step by step before concluding
  4. The human reads and checks each step, not just the conclusion

Pitfalls

Pitfall

Presenting this as a general accuracy booster for decisions overstates it; the source shows gains on reasoning benchmarks, not a guarantee for business decisions.

Sources

References

Listed A–Z by author surname, Harvard style. A source with a DOI links to https://doi.org/…; one without links to its page or says plainly that it has no DOI.

  1. Wei, J. et al. (2022) Chain-of-thought prompting elicits reasoning in large language models. (arXiv preprint arXiv:2201.11903).doi:10.48550/arXiv.2201.11903Source page