MINDPRINTBY AETHERISEnter console
Methodology

How a Mindprint is built

Observable professional writing and public business context enter an evidence ledger. Repeated signals become provisional interpretations, then practical meeting questions. Each interpretation retains its source references and its limits.

Applied intelligence

What the evidence helps you do before a meeting

Research priorities

Identify professional themes that appear in cited interviews, published writing and company materials.

Prepare questions

Turn source-backed observations into questions about business challenges without inventing personal motives.

Verify in conversation

Confirm what remains current, what the person actually means and where the available evidence was incomplete.

Mindprint is not a LinkedIn ranking optimizer, a validated personality test or a prediction of the next decision someone will make.

01 / MethodAnalysis pipelineThe ten stages in a full Mindprint analysis.Explore
  1. 01Intake
  2. 02Normalize
  3. 03Extract signals
  4. 04Cluster patterns
  5. 05Detect contradictions
  6. 06Calculate confidence
  7. 07Build behavioral model
  8. 08Goal-specific strategy
  9. 09Conversation & meeting prep
  10. 10Save Mindprint
02 / EvidenceConfidence scoringHow source quality, repetition and contradictions affect evidence confidence.Explore
contribution = base weight × source reliability × recency × observation type
claim score  = saturate(Σ contribution × repetition × cross-source agreement)
             − contradiction penalty   → clamped 0–100

recency half-life   365 days (floor 0.25)
observation weight  observed 1.0 · user-provided 0.8 · inferred 0.5
contradiction       −6 per record, max −30
< 4 records       cannot exceed "Moderate"

Labels: Emerging < 40 ≤ Moderate < 62 ≤ Strong < 80 ≤ Very Strong

Scores express model confidence and relative signal strength. They are not scientific measurements of a person.

03 / Safety and transparencyGuardrailsWhat Mindprint does not infer or claim, plus rights to export and delete.Explore
Observable professional evidence only

Only content the subject published professionally, or that you supplied, is analyzed. Every signal is traceable to its source.

Inference is not fact

Conclusions are hypotheses with confidence, evidence count and source references. Nothing is presented as certain.

No sensitive attribute inference

MINDPRINT does not infer race, ethnicity, religion, political affiliation, sexual orientation, disability, health or criminal propensity.

No diagnosis

This is not a medical or psychological assessment. Terms describe communication patterns, not personality or mental state.

No impersonation

Message Lab adapts structure and framing. It will not write as the person, and warns when output copies their unique phrasing.

No deception claims

Contradictions are shown as potential tension with possible interpretations — never as dishonesty.

Source traceability

Every claim opens an Explain drawer listing supporting and contradicting evidence.

Deletion and export

You can export or delete any Mindprint and its underlying source data at any time (Settings).

Weak evidence lowers confidence

Thin, old or single-source evidence caps confidence. Missing data is never filled in.

Dated coverage

Reports include scan date/time and evidence coverage.