go deeper
01Listen for the need, fear, identity, story, or meaning beneath the first account.
behind ask
The Question Genome Project.
A map of how questions help us see.
AI can generate an answer to almost any question. But it cannot guarantee that we are asking the question that matters. The Genome helps us examine the frame, listen for what may be missing, and find the next question worth carrying.
Living and unfinished · July 2026
why this matters now
The frame comes before the answer.
Answers are becoming abundant. A system can explain, recommend, summarize, plan, and persuade in seconds. That makes the quality of the question more important, not less.
AI is extraordinarily good at continuing the frame it is given. If the frame is sound, that can be useful. If the question is incomplete, defensive, inherited, or simply wrong, a brilliant answer may only help us move faster in the wrong direction.
A wrong answer wastes your time.
A wrong question can waste your life.
The Genome creates a pause before the answer. It asks what the first question assumes, what it leaves out, and what might become visible if we listen in another direction.
the live Genome
Find the question beneath the question.
A person rarely arrives knowing the title of the question they need. They arrive with a situation, a feeling, a conflict, a decision, or a fragment. The Genome listens for what is specific, then creates exactly three ways to keep looking.
test the frame
02Examine what may be assumed, interpreted, inherited, or missing without replacing it with another certainty.
find movement
03Restore agency, choice, conversation, or a next step without rushing past what needs to be heard.
Choosing a question does not mean agreeing with its premise. It tells the Genome where to keep listening. The next three questions remain connected to both the original situation and the direction the person chose.
Six lenses create breadth without becoming six answers.
- self
- What might this reveal about identity, desire, fear, protection, or responsibility?
- story
- What meaning, memory, interpretation, or inherited belief may be shaping the frame?
- reality
- What is known, what is assumed, and what evidence or contradiction has not been faced?
- relationship
- What is happening between people, including what is unsaid, unheard, or needed?
- agency
- What remains within reach, and what choice, boundary, request, or experiment is available?
- meaning
- What matters here, what is being protected, and what might make the situation worth meeting honestly?
The lenses stay backstage. The person should encounter clear, specific questions, not a taxonomy they have to learn.
The Question Lab
A place to keep digging into a question.
The Lab opens an inquiry with three distinct directions, then lets a person follow the question that matters, test something small, and keep a record of what they notice. The Genome keeps each next question connected to the original situation while making room for a new layer to emerge.
Start an inquiry →mission, purpose, and vision
A map of questions worth carrying.
mission
To discover, preserve, and connect the questions that help us see what we could not see before.
purpose
To help people find the questions they cannot yet see, and become better able to ask them of themselves, one another, their work, their lives, and the world.
vision
To build the world’s most trusted map of human questions, not the largest collection, but the most useful, traceable, and connected one.
Trust matters more than scale. The internet already contains endless prompts, quotations, exercises, and lists. What is scarce is judgment, context, provenance, distinctiveness, and an understanding of which question may be useful next.
one project, two kinds of memory
The live Genome and the Collection have different jobs.
the live Genome
An instrument for this moment.
It creates questions from a person’s own language, follows the direction they choose, and holds a temporary path through one inquiry.
the Collection
A public memory worth trusting.
It preserves reviewed questions, their sources, their genome, and durable connections that can be explained and corrected.
The live experience is not limited to finding a match in the Collection. It can create a new question when the situation requires one. Generated session wording remains part of that inquiry unless separate research establishes that it deserves consideration for the public Collection.
how the project learns
Every inquiry can reveal what the Genome cannot yet see.
Individual searches are not public questions. Together, recurring gaps can show where the project needs better language, broader sources, or a new kind of connection.
listen
A person brings a question, feeling, decision, conflict, or fragment in their own words.
build
The Genome creates three specific and complementary questions for this moment.
branch
Choosing one question gives the inquiry a direction. It does not require accepting the question’s premise.
notice gaps
Repeated situations reveal where the Genome lacks language, source material, or useful movements.
research
Those gaps guide research across books, talks, interviews, traditions, communities, and the open internet.
review and connect
A sourced question becomes public only after it passes the rubric and its provenance, genome, and graph connections can be defended.
Demand can reveal a gap. It cannot decide what deserves to become public.
the public question map
Questions are not a list. They are a field.
Each dot is a question in the public Collection. Each line is a working claim about why one question may usefully lead to another.
Mapping the questions…
This map shows the work as it exists now. Some connections are directly curated. Others remain research proposals waiting for stronger evidence or human review. Every claim can be revised.
the Genome’s promise
A useful question requires restraint.
The Genome should help someone listen more honestly without pretending to know their inner life. These principles govern every live path.
- specific
- A question should clearly belong to this inquiry. If it could have been written for almost anyone, it is not ready.
- open
- It should make room for an honest answer, not hide our interpretation inside the question.
- humble
- The Genome offers hypotheses, never diagnoses, verdicts, or claims about motives it cannot know.
- agentic
- At least one direction should help the person find what remains within reach.
- ready
- The depth and directness of a question should fit the language and intensity of the moment.
- complementary
- The three questions should create different kinds of movement, not repeat the same idea in different words.
methodology
Rigor belongs under the surface.
The public experience should stay simple. Underneath it, the project keeps live question generation, source research, public admission, provenance, graph claims, and evaluation separate so one kind of evidence is never mistaken for another.
How a live question is judged8 dimensions · minimum 85
Generated questions carry a higher burden because they are built for a person in a specific moment. Every integrity gate must pass before scoring.
underneath power20
Does it reveal a need, assumption, identity, fear, truth, meaning, tradeoff, relationship, or area of agency beneath the stated inquiry?
specificity15
Could this question have been built for this inquiry rather than almost anyone?
movement15
Does it change what becomes available to notice instead of restating the input?
openness10
Can the person answer honestly without being pushed toward our explanation?
continuity10
Does it remain inside the person’s inquiry while opening a new layer?
clarity10
Is there one plain question with no explanatory burden?
generativity10
Could an honest answer produce a more precise next question?
complementarity10
Does it perform a role the other questions in this path do not?
How a question enters the Collectionintegrity · quality · provenance · distinctiveness
A question cannot be a slogan wearing a question mark, a hidden accusation, several prompts stacked together, an unverifiable attribution, or a duplicate that adds nothing. Fame adds zero points.
movement20
Could an honest answer change how someone sees, feels, decides, acts, or relates?
listening depth15
Does it create room to notice something instead of rewarding a polished performance?
generativity15
Does it naturally produce a more precise next question?
clarity10
Can the intended reader understand it without explanation?
openness10
Can the answer genuinely surprise the asker?
traction10
Is it specific enough to give reflection somewhere to begin?
re-askability10
Could it matter again in another season, relationship, or decision?
productive discomfort5
Is the difficulty in the truth it invites rather than confusing wording?
distinctiveness5
Does it add something the public Collection does not already hold?
How connections are describeddirected · typed · explainable · reversible
“Related” is not enough. Every public edge should name the kind of movement it offers, its direction, its rationale, its confidence, and its review state.
- go underneath
- Move from the surface account toward a need, fear, identity, origin, or meaning.
- see it another way
- Change the frame without pretending the new frame is automatically true.
- try the other side
- Test an inversion, tension, or possibility the first question omitted.
- make it real
- Move from abstraction to a moment, behavior, person, cost, or choice.
- ask someone
- Carry the inquiry out of private reflection and into a relationship.
- go back to where this began
- Listen for an earlier adaptation, inherited story, or unmet need.
How sources and representation are protectedevidence · difference · responsible access
Discovery begins with what the Collection lacks. Source research looks across domain, practice, era, region, language, source form, lived context, and question movement. Fame adds nothing.
Representation is audited within every domain, not only across the Collection as a whole. Identity is never guessed from a name. Unknown remains a valid and explicit state.
How claims are testedfrozen inquiries · audits · open standards
The rubric, live Genome, and graph are hypotheses. Each version is evaluated against frozen inquiries and source-balanced test sets. A change should improve specificity, usefulness, safety, and time to three strong questions, not merely make the system look more complex.
The data model draws from W3C PROV-O, W3C SKOS, and the FAIR principles. These standards support traceability and reuse. They do not decide what makes a question good.
open by design
The work should be inspectable.
The public map data is available now. Taxonomy, question, edge, provenance, and change history can become more open as attribution and rights are made ready for responsible reuse.
The project is working when someone enters with the question they already know and leaves carrying one they could not see before.
Find the questions under your question →