We Do Not Teach Thinking to AI
Source: Dev.to
Overview
Most of us learned to prompt AI by guiding its thinking:
- “Think step by step.”
- “Here’s an example of how to solve this.”
- “First check A, then compare B, finally conclude with C.”
These techniques made sense because earlier models needed an explicit path; without structure they would rush to a conclusion.
Reasoning Models Shift the Premise
General conversational models excel at producing natural answers quickly. For tasks where the direction is clear—brief summaries, simple explanations—this is sufficient.
Reasoning models work differently. Rather than pushing problems straight toward conclusions, they are designed to:
- Compare conditions
- Trace possible paths
- Hold problems longer before forming answers
Models such as Claude’s extended‑thinking mode or OpenAI’s o‑series embody this direction, spending more computation on internal reasoning.
A reasoning model isn’t simply one that writes longer answers; it’s built to grapple with harder problems for a longer period.
Prompting Strategies for Reasoning Models
With general models, “think step by step” can be helpful because it forces intermediate steps rather than jumping to conclusions.
With reasoning models, the same approach doesn’t always work. When you strongly specify an arbitrary sequence of thinking to a model already designed to break problems down, you narrow the space for it to find a better path.
The same applies to examples. Good examples set the standard for an answer, but overly detailed examples can lock the model into a specific solution method—even when a superior approach exists.
This isn’t about Chain‑of‑Thought being wrong; it’s about using the same habits when your tool has fundamentally changed. With reasoning models, sometimes saying less is better.
What to Provide
- A clear goal
- The criteria for a good answer
- The output format you need
Then leave the middle steps to the model. The model doesn’t need you to design its thinking process; it needs to know what counts as a good answer.
This is an excerpt. The full piece—including a side‑by‑side prompt comparison and when reasoning models are the wrong tool entirely—is at Dechive.