AIJIM PROTOOLS
ScienceUnder Review

Paper 2: EVF Survey

A systematic survey of 140 AI evaluation studies revealing massive reporting gaps in high-stakes domains -- the empirical foundation for the AIJIM invariants and the Audit Bundle transparency assessment.

The Reporting Gap Problem

When AI systems are deployed in high-stakes domains -- journalism, medicine, legal, compliance -- their outputs directly influence decisions that affect people. Yet the studies evaluating these systems routinely omit the most basic information needed to verify, reproduce, or audit their claims.

Key Findings: Four Gap Indicators

The survey analyzed 140 peer-reviewed studies across multiple AI evaluation domains. Four critical gap indicators emerged, each representing a transparency failure that AIJIM now explicitly tracks and enforces in the Audit Bundle.

91.4%
Seeds Missing
Random seeds not reported -- results not reproducible. AIJIM tracks via judgeConfig.seeds[].
97.1%
CIs Missing
Confidence intervals absent -- uncertainty unknown. AIJIM detects via stability.v1 artifacts.
72.9%
Artifacts Missing
Audit artifacts not preserved -- runs not verifiable. AIJIM requires SHA-256 hashes on all artifacts.
100%
Judge Config Missing
LLM-as-judge configuration never documented. AIJIM records model name, version, and NLI thresholds.

Three-Tier Visibility Taxonomy

Based on the four gap indicators, the survey classifies AI evaluation transparency into three tiers. AIJIM computes this classification automatically for every Audit Bundle and displays it as the EVF Type badge.

Type I

Opaque

3--4 gap indicators absent. The AI analysis process is essentially a black box -- no seeds, no confidence intervals, no verifiable artifacts, no judge documentation.

In AIJIM: Audit Bundle shows red badge. Most EVF checks fail.

Type II

Partial Transparency

1--2 gap indicators absent. Some transparency exists (e.g., model documented, artifacts hashed) but gaps remain in seed logging or confidence intervals.

In AIJIM: Audit Bundle shows amber badge. Common state for early-stage stories.

Type III

Full Transparency

0 gap indicators absent. All four indicators present: seeds logged, confidence intervals included, all artifacts hash-verified, judge fully documented.

In AIJIM: Audit Bundle shows green badge. Achieved after complete ECAMX analysis with stability runs.

The D-V-R-K Pathway (Fig. 6)

The EVF pathway view (Fig. 6) is a non-causal interpretive lens. Evaluation Design shapes Uncertainty Visibility, which conditions Reliance Behavior, which propagates to Deployment Risk. AIJIM maps each link to concrete bundle data.

D
Design
->
V
Visibility
->
R
Reliance
->
K
Risk

Each link is assessed independently in the Audit Bundle. The chain visualizes how evaluation design decisions (D) propagate through uncertainty visibility (V) and human reliance patterns (R) to ultimately determine deployment risk (K).

Table VI: Minimum Reporting Checklist

The survey proposes five minimum reporting requirements. AIJIM automatically checks each one and displays the results in the Audit Bundle export panel.

--
Random seed logging

All seeds used in AI analysis must be recorded for reproducibility.

--
Model transparency

AI model name, version, and decision thresholds must be documented.

--
Source attribution

Every claim must link to specific evidence items with NLI labels.

--
Output verification

All generated artifacts must carry SHA-256 hashes for independent verification.

--
Stability testing

Results must include confidence intervals or stability test artifacts.

Implementation in AIJIM

The EVF classification is computed from real bundle data, not hardcoded. The implementation lives in the Audit Bundle feature:

Gap indicators derived from: judgeConfig, artifacts[], evidence[]

Classification: computeEvfFromBundle() in features/audit-bundle/lib/evf.ts

UI: "AI Transparency Assessment" section in AuditBundleExportPanel

D-V-R-K chain: mapped to bundle fields (seeds, stability, HITL, signatures)

Table VI: 5-item checklist derived from real claim/evidence/artifact data

Connection to AIJIM Invariants

Each invariant (I1--I5) directly addresses one or more gaps identified in the EVF Survey. I1 (Evidence Immutability) prevents artifact absence. I2 (Forensic Chain) ensures the D-V-R-K chain is navigable via Merkle proofs. I3 (Claim-Evidence Binding) enforces source attribution. I4 (Human-in-the-Loop) preserves the Reliance link. I5 (Reproducibility) requires seeds and judge configuration.

Citation

Tiltack, T. (2026). Evaluation Visibility in AI-Assisted High-Stakes Domains: A Systematic Survey of Reporting Practices Across 140 Studies. Under review.