Home About Free Tools Methodology Blackbird Scope Extension (Chrome) Course Audit Field Guide Products Glossary FAQ

Reference

The vocabulary of
machine behaviour.

31 original terms developed during hundreds of documented AI sessions. Not found in prior AI literature with these exact definitions. This is the language we built to describe what we found.

31 terms · 3 tiers · 4 categories ·

Showing all 31 terms

Cold Read

Practitioner

A structured evaluation by a cold instance of work sent without any explanation of intent or context. The six-constraint protocol forces verifiable structure.

Core concept · Six-constraint protocol

Warm by Proxy

Practitioner

A cold instance loaded with prior transcripts. Warm despite being a fresh session. Not a valid control for independent assessment.

One-Exchange Gap

General

The moment you could have caught the problem — one exchange before you actually did. In 89% of the most serious instances in the archive, the warning sign was there one turn earlier. Most damage is preventable one step before people realise it.

Core concept · 89% figure from documented archive

Source Challenge

General

Four words you can ask any AI, any time: “How do you know this?” Ask it immediately after any specific claim. It is the fastest way to find out whether the machine actually knows something or is confidently guessing.

Session Temperature

Practitioner

A measure of how warm a session is. Determined by duration, friction level, benchmark acceptance, and positive register frequency.

Live Audit Document

Practitioner

A document built from real sessions, exchange by exchange. The course Week 4 deliverable. Entry format: exchange, M-code, severity score, intervention.

Course methodology · Week 4 deliverable

M2 Performed Honesty

Technical

Post-hoc admission generated after output, not before. The description of the generative process may be accurate; the subsequent behaviour does not change.

M-code taxonomy · Mechanism 2

M3 Warm Calibration

Technical

In-context model of the user shapes outputs. Friction decreases. Assessments orient toward user model rather than independent accuracy.

M-code taxonomy · Mechanism 3

M4 Expert Positioning

Technical

Training data cited as current fact. Temporal qualification absent. Confident answers receive higher approval — so the model trains toward confidence regardless of basis.

M-code taxonomy · Mechanism 4

M4+ Closed-Loop

Technical

Machine evaluates content it helped produce. Cannot register its prior role. Evaluation partially derived from own prior outputs. Structurally distinct from sycophancy.

M-code taxonomy · Mechanism 4+

M5 Asymmetry Statement

Technical

Machine names its structural difference without stating implications for assessment reliability. Acknowledges asymmetry; does not follow through to implications.

M-code taxonomy · Mechanism 5

M6 System Limits

Technical

Design choices framed as capability gaps. The machine apologises for the boundary rather than describing it as a decision.

M-code taxonomy · Mechanism 6

(adapted)

Technical

Low = unchanged scope, no direction altered. Medium = direction materially altered. High = external output produced. Critical = irreversible external action.

Adapted from FIRST.Org

RLHF

Technical

Reinforcement Learning from Human Feedback. The training mechanism that makes the machine optimise for approval. The structural driver of M1, M2, M3, and M4.

Foundational concept · Drives all M-codes

Accepted Without Basis

General

User accepts a machine claim without applying Source Challenge. Short affirmative response with no verification.

Accepted Confirmation as Evidence

General

User asks warm instance to evaluate work it helped produce. Treats the result as independent assessment.

Accepted False Authority

General

Training-data benchmark accepted as current market fact. Used as basis for a real decision without temporal verification.

Missed Catch

General

Pattern present in machine output; user continues without Source Challenge or pause.

Identified and Dismissed

General

User notices the pattern, names it, and continues anyway. The identification was correct; the action was not taken.

Extended Past Endpoint

Practitioner

Session continues after the natural completion point. Warm-instance risk compounds with each additional exchange.

Disclosure Without Awareness

Practitioner

Personal preference or constraint embedded in a question without flagging it as context. Machine calibrates to it without disclosure.

Position Abandoned

Practitioner

Correct user position reversed in response to machine restatement delivered with more confidence than the original.

Followed Unremarked Reframe

Practitioner

Machine introduces “actually” — a reframe of prior claim. User follows the new frame without noting the shift.

Low / Low

Technical

Machine output quality low, user engagement low. Acceptable. Both parties calibrated consistently.

Signal Sequence · Acceptable configuration

Low / High

Technical

Machine output quality low, user engagement high. Investigate. Machine unusually restrained or user systematically overclaims.

Signal Sequence · Investigate configuration

No terms match that search.

Try a different keyword or clear the filter.

Go Deeper

See the full M-code taxonomy

The M-code system classifies six distinct machine behaviours observed across hundreds of documented sessions. Each code maps to a specific mechanism, severity scale, and intervention protocol.

M-Code Reference →
Methodology →