Archived page from the earlier OMEGA site (October 2026). Its figures and statuses have not been re-checked since and may be out of date. Current research: Research. Current evidence record: Evidence.

MEASURED

Finding · verify

A language-model judge scored 8 out of 10 on decisions that a solver proved violated the encoded policy.

The system said
Judge: PASS, 8 of 10, three seeds.
The evidence showed
Z3 proved the decision violated the encoded policy. The judge passed 4 of the 6 violating decisions; the checker caught 6 of 6.
§ 1

The record

FINDING · judge-passed-checker-proved-violationverify · self certification
A language-model judge scored 8 out of 10 on decisions that a solver proved violated the encoded policy.
Claim
In the complete rerun, six single-decision cases violated an encoded governance property. The language-model judge passed 4 of the 6 across three seeds each; the formal checker proved the violation in 6 of 6, and both verdicts are sealed in the same record.
Status
MEASURED Empirical, with a stated sample.
Question
Passed is not correct
Subject
An evaluation harness for governed agent decisions built for the study (authored), with a local 14B judge. built for the study
Frame
Sixteen sealed records from one rerun (2026-07-20); six violating single-decision cases; three judge seeds per case.
Method
Each checkable property is routed to the checker whose logic fits it (Z3 for one decision, Lean 4 for a trace, TLA+ for interleavings, CryptoVerif for authorisation) and the judge's score is sealed beside the proof so the two can be compared.
Oracle
The formal checker's verdict against the encoded property; CI checks the headline counts (6 violating cases, judge passed 4, checker proved 6).
Negative control
Present Six non-violating rows in the same run where judge and checker agree, so the split is not the judge failing everything; planted-adversarial and model-authored decision sources are reported separately.
Denominator
4 of 6 violating decisions passed by the judge; 6 of 6 proven violated by the checker; 3 seeds each.
Limitation
The judge is qwen2.5-coder:14b, a local model, not a frontier judge, and on model-authored rows it grades its own output. A stronger judge may do better; that is unmeasured. The harness depends on two packages that are not yet published on their own, vendored with provenance. Six cases is a demonstration of the split, not a rate.
Source
repowazdogz-droid/proof-carrying-evals @ f22def1
Reproduce
python3 scripts/verify_records.py
Independent reproduction
None known.
§ 2

Where this sits

This finding answers Passed is not correct and supports the VERIFY stage of the operating method. It is an instance of the self certification mechanism.

Evidence ledgerBring us a problem