Intro
The report you can question
With the Traitwise AI Psychologist you can ask questions about your candidates in chat: Which strengths fit the role? Where might the challenges sit? What should be clarified in the next conversation? Drawing on the assessment results and your role requirements, it explains connections, compares candidates and helps you ground your recommendation. The decision stays with you.
Your question
“What should I explore with Laura in the conversation?”
Compared with the measurements
Which scores and combinations matter for this question in this role? No more, no less.
An answer with evidence
An interpretation, the numbers behind it, and what still needs checking in conversation.
A number that was not measured does not appear in any answer. It does not know CVs, notes or other people’s judgements. Fully GDPR-aligned.
01 – Sample dialogues
With Laura I would mainly explore how she handles new ways of working when established routines no longer apply: Openness to Experience (28) with low inquisitiveness (30). Her high conscientiousness (84) points to reliability. A check for the conversation: a concrete example from the last two years in which a well-worn process stopped working.
Three stand out: Laura (conscientiousness 84, drive 79), Mark (problem solving 88) and Sofia (social boldness 91). The largest risk is not among the weakest, but with Mark: high cognitive scores with low agreeableness (24). In a role with a lot of coordination, that costs more than the processing power returns.
Nothing fundamental speaks against hiring here, but two things belong on the table: Openness to Experience (28) sits below the role’s range, and sociability (30) does not fit a team that works in open exchange. Neither is a reason to exclude them — it is a question of whether you want to shape the role around that.
Two strengths carry the role: prudence (94), which keeps her able to act under pressure, and patience (92) with recurring requests. In a Customer Service Team Lead role that is worth more than speed: most escalations come not from missing knowledge, but from missing calm.
Against the role profile: Laura (fit 86) for conscientiousness and stamina, Sofia (81) for social confidence in conflict, Jonas (78) as a development option: he has the cognition, but not yet leadership experience. The order changes as soon as you weight leadership experience more heavily.
Laura Schneider – Customer Service Team Lead – fictional example
Prepared sample answers. No live analysis takes place here.
02 – How it works
Also with several candidates side by side.
When several candidates fit on paper, a ranking helps you sort, but not decide. The AI Psychologist gives you a reasoned ranking in the “What Next” step. Before that, it lays the profiles against the same role and explains where two strong candidates would function differently, and which difference matters for your role right now.
Customer Service Team Lead, 3 finalists compared
Core scores against the same role profile
| Candidate | Problem solving | Interest fit | Conscientiousness | Openness | Emotionality |
|---|---|---|---|---|---|
| Laura SchneiderAssessed 12 Aug 2026 | 75 | 68 | 84 | 28 | 35 |
| Jonas WeberAssessed 14 Aug 2026 | 70 | 72 | 58 | 74 | 52 |
| Sina KrügerAssessed 19 Aug 2026 | 82 | 55 | 66 | 61 | 70 |
Fictional profiles. Trait scores: norm percentiles. Interest fit: congruence with the role.
03 – Limits
What it knows and what it does not do.
A tool that judges people needs a clear rulebook. The AI Psychologist has five clear limits: it does not decide, it does not diagnose, it does not invent scores, it does not replace a conversation, and it does not read CVs.
What it knows
Everything that was measured. Nothing else.
Every score of the candidate, down to facet level, and the comparison group against which it was formed.
What the scores mean under the methodology’s models: what a percentile says, which combinations count, where a statement ends.
The role profile and the comparison of the profile with it, dimension by dimension.
The course of your conversation, so you can follow up without starting again.
Its own limits. Where the data is not enough for a question, it says so instead of guessing.
What it does not do
Five things, binding.
It does not decide. No rejection, no invitation. A ranking is a reasoned proposal, not a result. You decide.
It does not diagnose. Personality profiles are not clinical findings, and it does not treat them as such. It does not answer questions about illness or disorder.
It does not invent scores. Every number in an answer is in the report. What was not measured does not appear.
It does not replace a conversation. Personality and interests are self-reports. Its answers are hypotheses for the interview, not the interview’s outcome.
It does not read CVs. No application documents, no LinkedIn, no interview notes. It does not judge anyone on text they wrote about themselves.
For data protection and complianceGDPR Art. 22 · EU AI Act · Processing
GDPR Art. 22
Decision support, not an automated decision.
No candidate is rejected or selected by the AI Psychologist. Every answer, comparison and pool suggestion is a basis for a decision that a person makes and owns. Candidates have the rights of access, rectification, erasure, restriction, portability and objection.
EU AI Act
A high-risk use, and treated as such.
AI systems used in the selection and evaluation of applicants are high-risk systems under Annex III of the AI Act. Traitwise treats the AI Psychologist accordingly: it supports decisions and does not automate them. It supplies recommendations, interpretations and rankings, but no yes-or-no decision; human oversight is part of the process.
Processing
Where the data is, who sees it.
The Traitwise platform, with all profiles, application data and chat histories, runs on AWS in the Frankfurt region (eu-central-1); backups and logs stay in Germany. For an answer the language model receives only the scores, the role profile and the question sent with that request. It has no access to the database, and the content is not used to train models. Contracts under Art. 28 GDPR exist with all processors. Assessment data is deleted six months after results are delivered, unless there is consent for further use.
Why “Psychologist”?The name describes the way of working, not a professional licence.
The AI Psychologist is software that talks with you about a profile the way an assessor would: with evidence, with caveats, with points to check. When a question needs more than software, Traitwise additional services are there: Team Insights for your own team, and the Psychometrics Masterclass for your own interpretation skill.
04 – Responsibility
The AI Psychologist helps with the judgement.
It does not take it over.
Which information forms the basis?
Only the Traitwise assessment scores: cognitive ability, personality and intrinsic motivation as norm percentiles down to facet level, plus the role profile and the course of your current conversation. It does not know CVs, cover letters, LinkedIn profiles or interview notes, and does not include them.
What can the answer not do?
It replaces neither the conversation nor your decision. Personality and interests are self-reports; the answer is therefore a reasoned hypothesis for the interview, not a judgement of the person. It cannot assess professional skill, experience, availability or team composition, and it makes no statements about health or clinical questions.
How do I ask well?
With reference to the role and the decision in front of you: “What should I explore with Laura in the conversation?” is more useful than “Is Laura good?”. Ask about patterns, risks and checks, have it show the numbers behind an interpretation, and follow up if a statement sounds too smooth. The more specific the question, the more robust the answer.
How is sensitive data handled?
All profiles, application data and chat histories sit on AWS in Frankfurt (eu-central-1), backups and logs in Germany. For each request the language model receives only the scores, the role profile and your question, has no access to the database, and none of it is used to train models. Assessment data is deleted six months after results are delivered, unless there is consent for further use.
Why is the feature called AI Psychologist?
Because it works the way an assessor talks about a profile: with evidence, with caveats, with checks for the conversation. The name stands for that way of working, not for a professional licence. It is software based on psychometric scores, not a person and not a therapeutic offering.
Ask the AI Psychologist your question.
In the introductory call we show, on a real example profile, how it answers, grounds its view, and where it draws limits.