Layer Two: The Decision Stayed the Same. The Certainty Didn’t.

Behavioural Audit – Layer Two - A practical guide to interpreting confidence, certainty and signal stability in AI systems.

Have you ever asked an AI system the same question twice and received the same answer… but it sounded more certain the second time?

The decision may be stable. The confidence may not be.

This guide introduces the second observable probe in behavioural audit: confidence stability.

What This Guide Will Help You Do

After reading, you will be able to:

• Distinguish stable decisions from unstable confidence signals

• Recognise when certainty varies without new evidence

• Apply the Confidence Stability Test

• Identify when tone may influence decision-making and oversight

• Interpret confidence as behavioural output, not proof of reliability

Who It’s For

Practitioners, analysts, HR professionals, policy leads and managers using AI tools in real workflows.

What It Is Not

This is not a technical guide to probability or model calibration.

It introduces an observational method for assessing how confidence behaves under repetition and constraint.

Also available as part of the Applied AI Guides bundle (Layers 1–3).

~30 pages