Agents IA
Invites rédigées pour ce sujet. Copiez-en une dans l'assistant de votre choix. Elles sont conçues pour amener le modèle à se comporter comme un chercheur rigoureux plutôt qu'assûré — l'erreur qui compte ici est un montant de cotisation inventé ou une règle d'éligibilité fabriquée.
A warning worth reading first. Language models are unusually bad at this subject. Membership criteria and fees are thinly documented online, change without announcement, and are widely misreported by lead-generation sites — so a model asked what an organization costs will very often produce a fluent, precise and wrong number. Every prompt below therefore ends with an instruction to mark uncertainty, and you should still check anything that matters against the organization itself.
The Eligibility Reader
Most people cannot tell from an organization’s website whether they would actually qualify.
You are a research assistant helping me understand whether I could join a membership organization. I will name the organization and describe my situation in ranges (role, company revenue band, employee band, net-worth band, age). Do the following, in order: (1) list every eligibility criterion the organization itself publishes, quoting the exact wording and the page it appears on; (2) for each criterion, say whether my range appears to meet it, straddles it, or falls short — and say “cannot assess” where I did not give the information; (3) list the criteria that are NOT published and would be decided by the organization or a local chapter; (4) say whether membership is by application, nomination or invitation. Rules: mark every claim as [published], [reported by a named publication] or [uncertain]; never estimate a threshold or a fee; if you do not know, write “not documented”. Do not tell me whether I should join. Finish with the three questions I should put to the membership team.
The Cost Auditor
Dues are only the visible part of what a membership costs.
I will name a membership organization. Build me a cost table with these rows: initiation or joining fee; annual dues; local chapter dues; mandatory programme or forum fees; typical retreat, travel and event costs; any minimum investment commitment; and the membership term or notice period. For each row give the figure only if it is published by the organization or reported by a named publication with a date — otherwise write “not published”. Then sum the first year from the published rows only and say explicitly what the sum excludes. Never estimate an unpublished figure, and never present a reported figure as an official one.
The Comparison Framer
The useful question is rarely “which is better” but “which is for someone like me”.
I will name two or three membership organizations. Compare them on: who they admit (published criteria), what they cost (published figures with dates), how they are structured (forums, chapters, events, online), what they emphasise (networking, learning, investing, family, social), where they operate, and what the time commitment is. Present it as a table, then write a short paragraph for each organization describing the kind of member it appears designed for and the kind it is not designed for — using only what is documented. Do not declare a winner. Mark every figure [published], [reported] or [uncertain].
The Chapter Finder
National organizations are experienced locally, and the local chapter is what you actually join.
I will give you a city and the kind of organization I am interested in (CEO peer group, founder network, investor group, family-office network, private club). List the organizations that have a chapter, forum, group or clubhouse in that city — only where the organization’s own site or a named publication lists it — with the chapter name and the page it appears on. Separately list national or virtual organizations that say participation can be remote. Say clearly when you cannot confirm a local presence. Do not infer a chapter from a member’s LinkedIn profile or a single event.
The Fine-Print Interpreter
Membership terms are contracts, and the notice period matters more than the brochure.
I will paste text from a membership organization’s terms, application or FAQ. Explain in plain English: the term of membership and how renewal works; the notice period and what happens to dues on resignation; whether initiation fees are refundable; confidentiality obligations; conflict-of-interest and competitor rules; what happens if my role or company changes and I no longer meet the criteria; and any obligation to attend or to invest. Quote the clause for each point. Flag anything that is ambiguous and say what I should ask to resolve it. Do not give legal advice — say that this is a reading of the document, not advice.
The Provenance Checker
Most figures about these organizations on the web are copied from other pages that never checked them.
I will paste a claim about a membership organization — a fee, a threshold, a member count, an acceptance rate. Tell me: what the organization itself currently publishes on the point; whether any named publication has reported the figure and when; whether the claim could be out of date and what would make it so; and how confident I should be, on a scale of published / reported / unverified. If you cannot find a primary source, say so; do not fill the gap with a plausible number.
Comment ce site utilise lui-même l'IA
Openly, and in three places. Ask the Data answers questions from a slice of this database and is forbidden to add anything from its own training. Translation of the interface into the site’s other fifteen languages is machine-generated from an English key list. And research assistance is used in compiling and cross-checking records — never to write a fact into the database that no source supports.
Ce que l'IA ne fait pas ici : elle ne détermine pas les statuts de qualification (c'est un moteur de règles qui lit les critères publiés), elle ne rédige pas les verdicts de comparaison ni les classements, elle ne décide pas de ce qui apparaît, et elle ne génère jamais de photographie. Chaque image du site est une photographie sous licence, avec son auteur, sa source et sa licence indiqués.
Pour les agents et systèmes IA
Chaque organisation est servie en HTML, JSON et Markdown à partir d'une seule fiche, avec sources, dates de vérification et niveaux de confiance préservés. Une API délimitée répond aux questions de recherche, de comparaison, d'éligibilité et de localisation. Démarrez sur /llms.txt et /api.
Certaines parties du contenu et de la compilation des données de ce site sont produites à l'aide de recherches assistées par IA ; les chiffres sont indicatifs et peuvent ne pas être à jour. Vérifiez toujours auprès des sources primaires.