Chain-of-Thought (CoT)

Chain-of-thought prompting is a prompting technique where you ask the AI to explicitly reason step by step before giving its final answer. The simple version: append "Let's think step by step" to your prompt and watch accuracy on multi-step problems jump dramatically. Why it works: when the model "thinks out loud" it gets more chances to catch its own errors and stay on track instead of jumping to a wrong conclusion. Modern reasoning models (o1, GPT-5 Pro, Claude Opus 4.8 with extended thinking) bake this in by default — they generate long internal reasoning traces before responding. You can still use explicit CoT prompts on non-reasoning models for a similar effect.