Skip to content

Security & Compliance

Recognise risks before they escalate.

In security-critical roles, trust determines access – to weapons, sensitive data or critical infrastructure. Yet risk traits rarely show in a CV and hardly ever in a job interview.

  1. CV
  2. Job interview
  3. Risk stays invisible

What you get

Traffic lights instead of gut feeling.

Triple Eye combines ITR and EED. The report has three levels: a summary with five traffic lights, a detailed evaluation per dimension and the interview report with reasoning for every statement.

Triple Eye · Report · SummarySample report · fictitious person

Overall aggravation tendency

Red

Red: The answers show a clear tendency to present oneself in a better light. The other traffic lights can only be interpreted to a limited extent – professional interpretation and clarification in conversation are needed.

Captures
The tendency to answer not according to one’s own attitude but in a way that promises a better result – usually a common, rather unintentional positive self-presentation.
Model
Aggravation scale
Assessment
ITR and EED
Evaluated by Dr. Jörg Prieler
Triple Eye · Report · Detailed evaluationSample report · fictitious person

Detailed evaluation · ITR

Stanine 1–9

Self-competence

  • Conscientiousness

    ← Works with little thoroughness, care and reliability.

    Works thoroughly, carefully, prudently and reliably. →

  • Resilience

    ← Copes poorly with difficulties, resistance and stress.

    Copes well with difficulties, resistance and stress. →

  • Personal responsibility

    ← Does not voluntarily take responsibility for the organisation and colleagues.

    Voluntarily takes responsibility for the organisation and employees. →

  • Credibility

    ← Does not describe experienced situations and facts accurately.

    Describes experienced and observed situations accurately and insightfully. →

Because of the red aggravation light, these scores can also only be interpreted to a limited extent. In the report: all dimensions, separated into ITR and EED, each with norm scale and a description of both poles.

Evaluated by Dr. Jörg Prieler
Triple Eye · Report · Interview reportSample report · fictitious person

Summary

The person describes their working day calmly and concretely. Many answers are consistently very positive – in line with the red aggravation light. The interpretation needs to be clarified in conversation.

Sample statements – not taken from the assessment

  • „I like to plan my working week in advance.“

    Answer

    rather true

    Confidence

    0.82

    Reasoning

    Says that she draws up a list for the week on Monday and adjusts it as needed.

  • „When something is unclear, I would rather ask once too often.“

    Answer

    fully true

    Confidence

    0.91

    Reasoning

    Answers very positively but gives no concrete example. Clarify in conversation.

    Uncertain – flagged in the report

  • „I stay calm even under time pressure.“

    Answer

    partly

    Confidence

    0.58

    Reasoning

    Describes both: calm with familiar routines, tension with short-notice changes. Contradiction flagged openly.

    Uncertain – flagged in the report

Evaluated by Dr. Jörg Prieler

How to read the traffic lights

  • Green: unremarkable
  • Yellow: borderline, clarify in conversation
  • Red: caution, professional interpretation needed

Only traffic lights are interpreted, never a single scale. No result is ever the sole basis for a decision.

The assessments in detail

ITR

Insider Threat Risk-Inventory

Purpose
Assessing the insider risk of people in security-critical roles, e.g. with access to weapons or data, in critical infrastructure or aviation.
What is measured
16 dimensions in two groups: ten empirically derived personality dimensions (incl. resilience, personal responsibility, conscientiousness, credibility) and six risk factors based on the Endrass model.
Result
Stanine per dimension and four traffic lights (risk of violence, faking tendency, overall picture, projection). Result valid for twelve months.
Duration
approx. 30 minutes, 210 statements
Languages
German, English, Hungarian, Spanish; further translations possible

EED

Early-Extremism-Detection

Purpose
Early detection of personality patterns associated with extremism.
What is measured
Dark triad model (narcissism, psychopathy, Machiavellianism) and the Marneros model.
Result
Two traffic lights in the Triple Eye report: risk of racism and dark triad personality.

Triple Eye

The bundle

Purpose
ITR and EED in one assessment, optionally with a short cognitive battery.
Result
Summary with five traffic lights, plus a detailed report with stanine per dimension, faking bar and indications from response speed.
Duration
approx. 45 minutes
Protection against faking
Normative and ipsative statements, random selection from a pool of statements.

Duration and figures refer to the classic assessment (questionnaire).

Protection against faking

Faking is captured separately.

A feature of the classic assessment: anyone trying to appear better than they are leaves traces.

Statement pool

Aggravation

Response speed

  • Statement pool

    Each person receives a random selection from a pool of statements.

  • Normative and ipsative

    Two response formats check each other.

  • Aggravation scale

    A dedicated scale captures whether someone exaggerates or plays things down.

  • Response speed

    Unusual response times are flagged as an indication.

Measurement model: rating scale model (IRT), backed by factor analyses.

What the conversation adds

Result and reasoning. Side by side.

The AI interview maps the answers back to the same items as the questionnaire and adds confidence and reasoning. Every result is supported by passages from the conversation – traceable for experts and for compliance documentation.

  1. Questionnaire
  2. Conversation
  3. Mapped back to item scores

You choose the format: a classic questionnaire or an AI interview.

Also available in classic form – as a questionnaire from IR&C

How AI interviews work in general → Use cases

Who is behind it

The conversation is new. The method is proven.

The assessments were developed together with experts from criminology, psychiatry, state security and mathematical statistics.

Dr. Jörg Prieler

Test and military psychologist, author of the ITR

  • ITR: developed by Dr. Jörg Prieler; manual 2015.
  • Presented in 2019 at the ICT security conference of the Austrian Military Counter-Intelligence Office in Fürstenfeld.
  • Selection of air traffic controllers for Austro Control.
All assessments and milestones →
  • Data processing in the EU
  • Developed in Austria
  • IR&CKolloqo – an assessment platform by IR&C
How we handle the data →

For experts

Questions before you start?

First, we clarify which areas are involved and which goals apply. Then comes the conversational design: tone, order, examples and, if needed, multilingual versions. Only then does it go live. Technically, little is needed – invitation links are enough, and participation works without an app.

Pricing for Security & Compliance is on the pricing page. View pricing →

Conversation has always been the better method. Now it scales.

Together, we’ll work out which assessment fits your needs.

Book a callOr: try the AI interview →