AI Quality Assurance
We check AI quality at multiple layers: automated checks on every analysis, structured testing of the instructions behind it, and your professional judgment as the final authority.
Multi-Layer Quality Checks

Before the AI writes a single word of analysis, it’s operating under a structured framework synthesized from CIGIE, GAO, IIA, ACFE, and other professional standards. Evidence quality principles — including relevance, reliability, sufficiency, and validity — shape how every piece of evidence is assessed.
This framework wasn’t generated by AI. It was designed by an oversight professional with 25+ years of investigation experience.
After every analysis, separate automated checks run, including faithfulness verification (are claims supported by evidence?), coverage verification (does it address all parts of the question?), and retrieval quality assessment (how relevant was the evidence found?).
These aren’t the same process that generated the analysis. They’re independent evaluations.
No analysis reaches your report without passing through a structured review flow. The system requires you to expand the analysis, check at least one citation against the source evidence, and record your professional judgment.
Every review action is timestamped. Every judgment is logged. This isn’t optional.
All automated checks feed into a single Quality Confidence level: Established, Probable, Possible, or Insufficient. Faithfulness and coverage scores drive the primary assessment. The logic is documented, not hidden.
Retrieval quality can lower confidence by one level but can’t drag a well-supported analysis to Insufficient on its own.
The instructions that guide the AI are versioned. Prompt changes run an automated check in CI against fixed test fixtures and are compared with a baseline.
Software engineering discipline applied to AI instructions. This is a regression check on structure, not a measure of accuracy.
Ongoing Testing
The instructions that guide Inqura's evidence evaluation are tested in a structured way, and we keep testing them.
We build test cases with known answers and check how Inqura's evaluation handles them. When we change the instructions, we test the change before relying on it. We are preparing a pre-registered evaluation of Inqura's analyses against a test case with known answers, and will publish the results as measured.
Practical Impact
The AI retrieves relevant evidence using hybrid search (meaning-based + keyword)
It assesses each piece of evidence for relevance and material limitations
It produces a structured finding with citations and a confidence level
Automated quality checks (faithfulness, coverage, relevancy, retrieval quality, contradictions) run in the background after the analysis is generated
Results appear in your quality metrics panel when the checks finish
You see the confidence level, faithfulness score, coverage score, and retrieval quality
You can expand “Evidence Considered” to see exactly what the AI looked at
You review the analysis, check citations, and record your judgment
State changes and your recorded judgment are logged in the audit trail
Report sections are drafts built from your completed analyses. Review and edit each one before you export.
Limitations
This is why the investigator review flow exists. AI checks verify the mechanics. Your expertise verifies the substance. Security, data handling, and certification questions are answered in the Trust & Security FAQ.