Confidence score
A 0-to-1 value reflecting how certain the detector is that a piece of text was AI-generated.
A confidence score is a numerical value, typically between 0 and 1, that reflects how certain an AI detector is that the input text was generated by a language model. A score of 0 means "definitely human," 1 means "definitely AI," and 0.5 means "could go either way."
Different products surface this score differently. AI Detector API returns the raw 0-1 score plus a categorical confidence field (high / medium / low) so you can apply your own threshold or fall back to the categorical label.
Choosing the right threshold is a product decision, not a technical one. Schools may treat 0.6 as actionable; moderation pipelines often only act on 0.95+. The right number depends on the cost of a false positive vs. a false negative in your context.
Related terms
Move from definition to code
Free 1,000 requests/month — no credit card. Be detecting AI text in 5 minutes.