GFAB-2026-FOUNDING · demonstration only
Professional judgment profile
Amina Rahman
Fictional name and synthetic evidence · issued values would be bound to a sealed score snapshot
72
Expert-Referenced Judgment Index
Synthetic demonstration value. It is neither a population standing nor an award threshold.
Evidence state
Founding expert-referenced diagnostic
This sample shows how a founding report labels evidence and limits. It does not stand in for validation data.
Issued record
Completion attestation
A live record can confirm completion and current validity. This sample has no active verification state.
How to read this report
Diagnosis, not verdict
An issued GFAB report describes how the response set approached the situations in one governed edition. It does not certify character, future conduct, product knowledge, or universal capability.
Values remain attached to achieved and available evidence. Unavailable comparison evidence stays unavailable. Reasoning observations can disappear entirely when coverage, stability, confounding, or review gates are incomplete.
Evidence-led summary
The clearest reading
Across this synthetic response set, the most consistent evidence appeared in Professional Judgment and Communication & Trust, while Trade-off Recognition was meaningfully less settled.
Strength 01 · PJ
Protecting the client under pressure
Relative to the other dimensions in this synthetic response set, client protection remained most consistently central when the easier course could have displaced it.
Strength 02 · CT
Candour that preserves participation
Difficult information was generally handled without removing the client’s ability to understand and take part in the decision.
Focus · TR
Where the evidence was less settled
The synthetic responses varied most when a desirable gain needed to be weighed against lost flexibility or protection.
Eight-dimension reading
The complete judgment profile
Each dimension contains nine equally weighted judgment items. The display is descriptive and uses synthetic values solely to demonstrate hierarchy, density, and report language.
PS
Priority & Sequencing
69
descriptive index
The synthetic response set was aligned in some situations involving immediate protection and longer-term planning, while other situations produced a less settled sequence.
6.2 achieved evidence points · 9 available · complete synthetic evidence
RR
Risk Recognition
76
descriptive index
The synthetic responses consistently distinguished consequential, difficult-to-reverse exposures from visible but recoverable movement.
6.8 achieved evidence points · 9 available · complete synthetic evidence
CR
Client Reality
71
descriptive index
Client commitments and structural constraints generally remained inside the decision, with some variation when a generic planning rule looked attractive.
6.4 achieved evidence points · 9 available · complete synthetic evidence
BJ
Behavioral Judgment
68
descriptive index
The response set often preserved client participation while emotion was active, although this was not equally settled across every situation.
6.1 achieved evidence points · 9 available · complete synthetic evidence
TR
Trade-off Recognition
63
descriptive index
The evidence was least settled when an appealing benefit needed to be considered together with the flexibility or protection it displaced.
5.7 achieved evidence points · 9 available · complete synthetic evidence
SP
Scenario Planning
72
descriptive index
The synthetic responses usually followed a decision beyond its immediate effect into later dependencies and constraints.
6.5 achieved evidence points · 9 available · complete synthetic evidence
CT
Communication & Trust
79
descriptive index
Candour and client participation were consistently preserved when the situation required difficult or unwelcome communication.
7.1 achieved evidence points · 9 available · complete synthetic evidence
PJ
Professional Judgment
81
descriptive index
Client protection remained central when convenience, pressure, or conflict could have pulled the decision elsewhere.
7.3 achieved evidence points · 9 available · complete synthetic evidence
Confidence evidence
Confidence and reference alignment
0.180
Probability-quality score
72%
Mean confidence
75%
Reference-aligned selection rate
−3 pp
Raw confidence gap
In this synthetic profile, stated confidence and best-available expert-reference alignment were closely aligned. The evidence state comes from a versioned probability model rather than the raw gap alone. Binary practice assertions remain a separate evidence class.
Response-process evidence
Reasoning observation withheld
This synthetic report intentionally demonstrates suppression. No directional reasoning description is shown because the fictional evidence does not clear every predefined coverage, stability, review, confounding, and incremental-validity gate.
Withholding a pattern means “not sufficiently supported.” It does not mean that no pattern exists, and it does not create a neutral midpoint.
Comparison evidence
Reference comparison unavailable
This founding sample does not display a rank, geographic standing, role comparison, or assumed midpoint.
A future comparison can appear only when the cohort definition, sample, inclusion rules, uncertainty method, as-of date, methodology version, and independent approval are bound to a frozen snapshot.
Method and limits
What the founding diagnostic establishes
Method summary
Single-choice situations use governed 0–3 option rubrics. Ranking situations score only material precedence pairs. The eight dimensions carry equal weight. Confidence and suppressible reasoning evidence remain separate layers.
Declared limits
- This sample describes fictional evidence, not a real response set.
- The founding method does not establish a population standing.
- Reasoning output remains absent unless every evidence gate is met.
- One annual response set does not establish character or future conduct.
Provenance and validity
No live verification for a sample
Issued artifacts carry a view-model hash, score-snapshot reference, sealed source hash, artifact checksum, version, and current validity state.
A changed score, narrative, benchmark, consented identity field, or validity state creates a new version rather than silently rewriting the prior artifact.
Sample state
Not verifiable
view-model-hash: synthetic-preview-only