Skip to content
BENIBIX Learn

Refusal (AI)

Also known as AI Refusal, Model Refusal, Over-Refusal

When an AI model says no to a request. Most often it could cause harm.

Think of it like

It is like a shop worker who will not sell knives to a child.

Example

A nurse asks a model for the safe top dose of a common pill. A model that is too careful says no, though the answer is on the box.

Why it matters

Too few refusals let harm through. Too many make the model useless and push people to worse tools.

Definition

A refusal is when an AI model declines to do what was asked. Models are trained to refuse requests that could cause serious harm, such as help with weapons. This comes from training, not from a filter added on top. But refusals go wrong both ways. A model may refuse a safe request that only sounds risky, which is called over-refusal, or help with a harmful one that is worded cleverly. A good refusal says why, and offers a safe path where it can.
Two refusals, one right and one wrong. The wrong one is over-refusal.

Common mistake

Thinking more refusals means more safety. Saying no to safe questions is a failure too.

Key takeaways

  1. 1
    Trained in. Part of the model, not a filter.
  2. 2
    Two ways to fail. Too many, or too few.
  3. 3
    Explain the no. Offer a safe path.

Compare