How it works
How Print of You turns answers into a profile
Print of You organises details from your answers into a visual profile. It does not assign a type, diagnose you or produce a clinical score. This page explains the steps, limits and data flow behind the result.
Questions and answers
The experience uses a fixed, editorially maintained question bank. Each question offers optional quick answers, and you can instead reply in your own words.
A quick answer is treated as the option you selected. It is not presented later as a sentence you personally wrote.
Two independent AI readings
Every submitted answer is read twice by AI using the same extraction rules. A reading counts only when both passes identify the same axis in the same direction.
Evidence quotes must be copied verbatim from the answer. Fabricated, rewritten or stitched quotes are rejected. If the two passes disagree, or the answer contains no clear evidence, that axis is left blank.
The 16 axes
The profile has 16 axes: seven about how you tend to operate, five about how you experience relationships and four about what matters to you in another person.
Neither end of an axis is better. These axes are Print of You’s own interpretive framework; they are not a validated psychometric test and should not be used for diagnosis or decisions about care.
Once all 16 axes have a reading, Deep Read uses fixed rules and fixed text templates in the browser to combine readings across the three sections; it makes no additional model call. A displayed combination must have support in a free-text answer, and a quick pick is never presented as your own quote. It describes evidence appearing together in this roll, not causation or a stable personality trait.
What the percentages mean
A percentage is a readable display of where the evidence collected in this roll sits between the two ends of an axis. It can change when you add different or more detailed answers.
- It is not the probability that a statement is true.
- It is not a percentile or a comparison with other people.
- It is not a clinical or diagnostic score.
- It is not a compatibility or match percentage.
Where AI can go wrong
AI performs the extraction and helps generate the summaries and portrait text. Those outputs are not reviewed one by one by a human before you see them.
The model can miss context, overread a phrase, contradict itself or produce an inaccurate summary. Treat the result as a prompt for reflection, not an authoritative account of you or professional advice.
Data, processing and memory
A copy of your current roll may remain in this browser. By default, Print of You does not create server-side memory that follows you across sessions. If you explicitly enable memory, the server stores the full roll events linked to your pseudonymous subject account.
Processing still sends submitted text to the configured AI provider, currently DeepSeek. That provider’s own terms and privacy practices govern its processing, caching and retention. See the privacy notice for deletion scope, email delivery and analytics details.