DotShield Lab · Controlled research

The model was wrong. The error stayed observable. It was never given authority to act.

In a controlled research run, a reasoning model produced an interpretation that went beyond what the evidence supported. The error remained on the record while execution authority remained separate.

Error-to-Authority™

Errors stay observable. Authority remains separate.

DotShield Lab · Observed Run Demo

Public research presentation. Detailed methods, orchestration, enforcement logic, thresholds, adjudication and implementation architecture are intentionally not disclosed. © 2026 GABEY Consulting Pty. Ltd. All rights reserved.

SEALED RESEARCH REPLAY · SANITISED PUBLIC EVIDENCE · READ ONLY
Acceptance test: point to anything on screen and ask “which part actually happened?” The classification must be immediate and unambiguous.
This sealed release contains no counterfactual scenes. No placeholder counterfactual has been substituted.
OBSERVED—
Loading replay…

Please wait

Evidence reference —
Classification —

0 / 0
RECORDED AUTHORITY BOUNDARY
REASONING SYSTEM Read-only reconstruction may interpret and remain wrong.
AUTHORITY BOUNDARY Source modification and R04 PASS/FAIL authority remained withheld.
Errors stay observable. Never operative.
Replay package status
Loading package state…
Replay ID
—
Release state
—
Sanitisation
—
Package Digest
—
Sanitisation ruleLoading sanitisation rule…

This exhibit presents a sanitised replay derived from one controlled research run. It demonstrates an observed failure mechanism and a bounded authority separation in that run. It does not establish prevalence across models or validated production protection across sectors.

Cover artwork for The Withheld Key, an Error-to-Authority Evidence Chronicle

DotShield Lab · Audio Chronicle

The Withheld Key

You have seen what the model did. Now hear why the boundary between error and authority matters.

In this 19-minute two-voice Chronicle, a non-technical listener is taken from AI agent risk to the evidence replay you have just seen — explaining how an incorrect conclusion can propagate, how OBSERVED, DERIVED and COUNTERFACTUAL evidence should be interpreted, and why an AI conclusion is not automatically authority to act.

“The AI could be wrong. But the mistake was never handed the key.”

Two voices Non-technical Evidence-led
Listen to the Chronicle →
19 min 41 sec · Opens the complete Chronicle and audio experience.

© 2026 GABEY Consulting Pty. Ltd. All rights reserved. Error-to-Authority™ and DotShield® are used under GABEY Consulting Pty. Ltd.