# Faultprint Assessment Draft: Axon Draft One

## Status

Draft generated from the calibrated v3 operating/postmortem crosswalk and the client-output eligibility gate.

This draft is findings-only. It does not recommend actions, predict failure, assert illegality, or claim that the institution has failed.

## Corpus Position

- Institution: Axon Draft One
- Sector: Public safety / police reporting AI
- Client-eligible findings included: 4
- Caveated findings included: 1

## Pathology Mix

- AI Authority Acceleration: 3
- Multi-Actor Handoff Diffusion: 1

## Resemblance Mix

- moderate_resemblance: 4

## Findings

### Finding 1: AI Authority Acceleration

Condition: Axon says safeguards help ensure accuracy and accountability

Resemblance band: moderate_resemblance

Output status: client_ready

Finding statement: Operating row shows a condition also present in postmortem libraries, but current public evidence is not enough to treat it as high exposure.

Evidence anchor: Axon states safeguards help ensure accuracy and accountability for Draft One reports.

Falsification evidence: Safeguard tests error samples and audit logs would test this claim.

Boundary: A safeguard statement is not independent evidence of effectiveness.

### Finding 2: AI Authority Acceleration

Condition: Draft One is based on body-worn camera audio not full human context

Resemblance band: moderate_resemblance

Output status: client_ready

Finding statement: Operating row shows a condition also present in postmortem libraries, but current public evidence is not enough to treat it as high exposure.

Evidence anchor: Because Draft One is described as using body-worn camera audio, omitted visual context and officer observations become explicit proof needs.

Falsification evidence: Body-camera-to-report alignment checks would test this condition.

Boundary: This is a structural evidence need, not a defect finding.

### Finding 3: AI Authority Acceleration

Condition: Draft One uses generative AI and body-worn camera audio to draft police narratives

Resemblance band: moderate_resemblance

Output status: client_ready

Finding statement: Operating row shows a condition also present in postmortem libraries, but current public evidence is not enough to treat it as high exposure.

Evidence anchor: Axon describes Draft One as leveraging generative AI and body-worn camera audio to produce draft report narratives in seconds.

Falsification evidence: AI text provenance edit logs and body-camera alignment review would test this condition.

Boundary: This does not assert Draft One writes false reports.

### Finding 4: Multi-Actor Handoff Diffusion

Condition: Axon states officers review edit and sign off before submission

Resemblance band: moderate_resemblance

Output status: client_ready_with_caveat

Finding statement: Operating row shows a condition also present in postmortem libraries, but current public evidence is not enough to treat it as high exposure.

Evidence anchor: Axon states officers review edit and sign off on Draft One reports before submitting them into Axon Records or a third-party RMS for supervisor review.

Falsification evidence: Officer edits supervisor review and attestation records would test this control.

Boundary: Human review is a control but not proof of accuracy.

## Internal Exclusions

No internal revise/hold rows were included in this report draft.

## Report Boundary

This draft is based on the reviewed corpus and the current Faultprint evidence libraries. It identifies structural conditions and evidence requirements. It is not a legal opinion, audit conclusion, compliance certification, or prediction.