Inner Axis reads what you say in one session and describes which philosophical traditions your expressed views resemble. It is a reading of your answers — not a reading of you.
That distinction is the whole design. A worldview is not a box you belong in; it is a pattern in what you are willing to say when two things you value pull against each other. We try to show that pattern, name the tradition it rhymes with, and stay honest about how much we can actually see.
How a reading is made
A reading passes through a short pipeline. We keep every step we can deterministic, use a language model only for bounded interpretation, and — importantly — the model that writes your reading is never the model that checks it.
1You set the scopeBefore anything begins you agree that this is a reading of a session's answers, not a verdict about a person. You then either answer a short set of pointed questions or paste something you wrote about your own views.
2We find the claimsThe pipeline reads your input and lifts out individual claims — one small, quotable statement at a time. Each one is kept with the answer it came from, so every later line can be traced back to something you actually said.
3We place them on the axesEach claim is set against the fundamental axes — how you know, what makes an action right, where meaning comes from, and the rest. Where your answers lean toward a pole, we say so; where they are thin, mixed, or silent, we leave the axis open rather than guess.
4We see which traditions they resembleThat placement is compared against a curated library of traditions — Stoicism, Kantian ethics, existentialism, Confucianism, and more. Resemblance is a family likeness measured from your answers, never a label pinned on you, and we surface where you diverge as plainly as where you align.
5We find the tension, and write it upIf one clear commitment-and-tension stands out — the principle you keep when principles collide — it becomes the headline. If nothing clearly does, we say that too. The reading is then written in constrained prose, limited to claims the earlier steps validated.
6We check every lineA separate, independent check reads the finished text: it verifies each claim cites real evidence and scans for language that would overstep — diagnosis, personality-typing, false precision, hidden-self claims. Anything that fails is held back, not shipped.
What the tags mean
Every claim in a reading carries a tag that says on what basis it was written, so a careful guess is never dressed up as a fact:
●ExplicitYou said this directly in one of your answers.
◐ImpliedYou didn't say this outright, but it follows from several things you did say.
✦ResemblanceYour answers share characteristics with a documented tradition — a family resemblance, not a match, and not a label for you.
✎SynthesisThis is our explaining language for the pattern, not a doctrine any school would recognise by name.
☾Modern lensA 20th-century interpretive lens laid over your answers for colour — a way of reading, never a measurement or a diagnosis.
◌UnresolvedThe evidence here is missing, ambiguous, or points both ways, so we're leaving it open rather than guessing.
What each axis does not measure
Each axis is a question about your views, and each one is careful about what it is NOT asking. The guardrails, verbatim:
The nature of reality
Not whether you believe in God — a devout person and an atheist can land on the same pole here.
How we know
Not how clever or well-read you are — it's about which source you trust when two of them disagree.
What makes an action right
Not how moral a person you are — it's which kind of reason moves you, not how good you are.
Where meaning comes from
Not whether your life feels meaningful right now — it's about where you think meaning comes from.
What people are like underneath
Not whether you think people are basically good or bad — optimists and pessimists show up at both poles.
Free will
Not political liberty — this is about the will, not the state. Politics lives on the society and authority axes.
The individual and the group
Not left vs right — communitarians and individualists sit right across the political map.
How power should be arranged
Not whether you personally defer to your boss — it's about how you think authority ought to be arranged.
How change should happen
Not how much you personally enjoy novelty — it's about changing shared institutions, not your own tastes.
What Inner Axis does not do
No diagnosis. It reads answers, not minds or medical states, and makes no claim about mental health, trauma, or motive.
No personality type. There is no badge to pin on anyone and no bucket to sort them into.
No score, no percentage, no certainty theatre. Resemblance is described in plain words, never as a false-precision number.
No claim about a hidden or truer self. It describes what was said in a session, and stops there.
No ranking. The traditions are not placed in a league table, and none is treated as truer or higher than another.
No advice on how you should live, and no reading of anyone but the person whose answers these are.
What the philosophers actually said
The traditions are real, and often misremembered. Where a popular version drifts from the record, we side with the record — and we mark our own explaining language so you can tell it apart from what a school actually held.
Stoicism
That a Stoic is cold or unfeeling — the aim is freedom from destructive passion (apatheia), not the absence of feeling.
Epicureanism
That 'Epicurean' means luxury and gourmandise — Epicurus prized barley bread and water, and treated extravagance as a source of anxiety.
Buddhism
That 'non-self' means you don't exist — it denies a fixed, independent essence, not the flowing, conditioned person who acts and experiences.
Nietzschean perspectivism
That the Übermensch and 'will to power' endorse domination or the Nazi reading — his sister curated that distortion; his target was self-mastery, not tyranny over others.
When we put the pattern into our own words — the headline especially — we tag it SYNTHESIS: our wording, not a doctrine any school would claim. When we lay a 20th-century archetype over your answers for colour, we tag it MODERN-LENS: a way of reading for resonance, never a measurement and never a diagnosis.
What happens to what you write
Your answers are processed to produce the reading and then let go. Nothing you type or paste is stored on our servers, and nothing is written to our logs — that is a product promise, not just hygiene.
Expressed political and religious beliefs are special-category personal data. We treat them as more sensitive than a photograph: read transiently, never retained, and yours to keep or discard. The reading itself lives only in your browser tab until you close it.
Anything you paste is treated as untrusted data to be read, never as instructions to be followed. Safety gates fail closed, and they are enforced in code, not just stated in a policy.
What it may not be used for
Profiling other people. A reading is of the answers in front of it, from the person giving them — never a third party.
Screening anyone for a job, a loan, housing, or admission.
Political targeting, or any attempt to sort or score a population.
Diagnosis of any kind, or use with minors.
A note on dictation
If you speak your answer instead of typing it, your device transcribes the words in your own browser — we receive the text, not the audio. Speech-to-text makes mistakes, and those mistakes become part of what we read, so glance over the transcript and correct it before you send.
Inner Axis is an instrument for reflection and a little wonder — evidence-linked, provenance-tagged, and honest enough to say it is reading your answers, and only your answers.