Chain-of-thought prompting was discovered in 2022: simply adding 'Let's think step by step' to a math or reasoning question dramatically improved LLM accuracy. The mechanism: forcing the model to produce intermediate reasoning tokens gives it more 'compute' to work with — each generated reasoning step lets the model accumulate insight. By 2024, CoT was built into 'reasoning models' (OpenAI o1, DeepSeek R1) that automatically apply extended reasoning before answering. The trade-off: more output tokens (higher cost, slower). For complex problems (math, logic, planning), CoT often turns a 30%-accurate model into 80%-accurate. For simple tasks, it's overhead.
GLOSSARY
What is Chain-of-Thought (CoT) Prompting?
Asking an LLM to 'think step by step' before answering — significantly improves performance on reasoning tasks.
RELATED TERMS
Prompt Engineering
The practice of crafting input text (prompts) to get the best output from an LLM. A vague prompt gives vague answers; a structured prompt gives precise ones.
LLM (Large Language Model)
An AI system trained on massive text datasets to predict and generate human-like text — the technology behind ChatGPT, Claude, Gemini, and most modern AI chatbots.
Back to the AI Glossary