Artifacts · 07

Confidence and Abstention Statement

The Confidence and Abstention Statement records what Onepane concluded, at what confidence, what evidence was missing, and what could not be determined, because a confidently wrong root cause is worse than none.

Who it's for: Sceptical SRE and platform leads; regulated accounts where human sign-off is mandatory.

Contents

What does the Confidence and Abstention Statement contain?

The highest-trust artifact in the suite. Saying 'we don't know' in writing is what convinces a sceptical SRE that this is not another vendor overselling.

01 Confidence level for the primary conclusion
02 Evidence that would raise confidence, and where it lives
03 Explicit abstentions: 'insufficient evidence, here is what's missing'
04 Human reviewer sign-off
FAQ

Confidence and Abstention Statement, questions.

What if the AI is confidently wrong?

Abstention is a first-class output. When evidence is insufficient the report says so and lists what is missing, rather than guessing. Every claim links to evidence, human sign-off is mandatory, and the abstention rate is published monthly.

What is abstention in AI root cause analysis?

Abstention is the system's explicit decision not to state a root cause when the available evidence does not support one, with a written account of what is missing. It is the difference between decision support you can audit and a tool that always produces an answer.

Send us your last 90 days of Sev1s.We'll show you what we would have found.

Not a demo. A replay on your own incidents, scored against the RCA a human actually wrote. Two weeks, no cost.