{
  "version": "sha256-p7R6P9OtVAqL0S5j9Cqzwh66hu/XjI2G9cUugkXPPX4=",
  "dateModified": "2026-10-18",
  "about": "The approved claims behind /cr, one sentence each, with source and validity. A Node's content in /cr is its claims joined by one space.",
  "entities": [
    {
      "id": "the-graph-web",
      "kind": "Project",
      "url": "https://thegraphweb.com/#the-graph-web",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        },
        {
          "kind": "hasProblem",
          "target": "problem"
        },
        {
          "kind": "investigates",
          "target": "hypothesis"
        },
        {
          "kind": "hasVision",
          "target": "vision"
        },
        {
          "kind": "implementedBy",
          "target": "architecture"
        },
        {
          "kind": "exposesAction",
          "target": "contact"
        },
        {
          "kind": "exposesAction",
          "target": "newsletter"
        }
      ],
      "claims": [
        {
          "id": "the-graph-web.definition",
          "predicate": "definition",
          "text": {
            "en": "The Graph Web is an experiment with two parts. One is a Canonical Representation: the project's knowledge, written for AI agents. It is published as JSON at thegraphweb.com/cr and as a plain web page at thegraphweb.com. The other is AISwise, the chat at thegraphweb.com/chat, where people ask about the project.",
            "es": "The Graph Web es un experimento con dos partes. Una es una Representación Canónica: el conocimiento del proyecto, escrito para agentes de IA. Se publica como JSON en thegraphweb.com/cr y como página web sencilla en thegraphweb.com. La otra es AISwise, el chat de thegraphweb.com/chat, donde la gente pregunta por el proyecto."
          },
          "url": "https://thegraphweb.com/#the-graph-web.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "the-graph-web.principle",
          "predicate": "principle",
          "text": {
            "en": "AISwise answers only from that Canonical Representation. So what people can learn here is just what agents can read there.",
            "es": "AISwise responde solo a partir de esa Representación Canónica. Así, lo que la gente puede saber aquí es justo lo que los agentes pueden leer allí."
          },
          "url": "https://thegraphweb.com/#the-graph-web.principle",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "problem",
      "kind": "Problem",
      "url": "https://thegraphweb.com/#problem",
      "relationships": [
        {
          "kind": "motivates",
          "target": "hypothesis"
        }
      ],
      "claims": [
        {
          "id": "problem.definition",
          "predicate": "definition",
          "text": {
            "en": "More and more, AI agents read and act on digital systems that were written for people.",
            "es": "Cada vez más, los agentes de IA leen y usan sistemas digitales que se escribieron para personas."
          },
          "url": "https://thegraphweb.com/#problem.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "problem.detail",
          "predicate": "detail",
          "text": {
            "en": "Structure, such as what a piece of content links to and how, is only in the prose. An agent has to guess it, as it cannot look it up.",
            "es": "La estructura, como con qué se relaciona un contenido y de qué forma, solo está en el texto. Un agente tiene que deducirla, porque no puede consultarla."
          },
          "url": "https://thegraphweb.com/#problem.detail",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "problem.motivation",
          "predicate": "motivation",
          "text": {
            "en": "This is not just theory. It comes up again and again when you build AI products and automations: the same hidden structure has to be worked out from the prose each time, when it could be looked up once.",
            "es": "No es solo teoría. Aparece una y otra vez al crear productos y automatizaciones con IA: la misma estructura oculta hay que sacarla del texto cada vez, cuando se podría consultar una sola vez."
          },
          "url": "https://thegraphweb.com/#problem.motivation",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "hypothesis",
      "kind": "Hypothesis",
      "url": "https://thegraphweb.com/#hypothesis",
      "relationships": [],
      "claims": [
        {
          "id": "hypothesis.definition",
          "predicate": "definition",
          "text": {
            "en": "This project tests one idea. Can AI agents understand and act on digital systems better when their knowledge is a graph? Here that graph is called a Canonical Representation. The other choice is text written for people.",
            "es": "Este proyecto pone a prueba una idea. ¿Entienden y usan mejor los agentes de IA los sistemas digitales cuando su conocimiento es un grafo? Aquí ese grafo se llama Representación Canónica. La otra opción es texto escrito para personas."
          },
          "url": "https://thegraphweb.com/#hypothesis.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "hypothesis.evidence",
          "predicate": "evidence",
          "text": {
            "en": "A smaller test on format alone found no gain it could measure. In July 2026 we wrote the test plan first, then gave an AI model the same facts as a document and as a graph. It scored the same both times: 8 of 8 tasks, both to recall facts and to take an action.",
            "es": "Una prueba más pequeña, solo sobre el formato, no encontró ninguna ventaja medible. En julio de 2026 escribimos primero el plan de la prueba y luego dimos a un modelo de IA los mismos datos como documento y como grafo. Sacó lo mismo las dos veces: 8 de 8 tareas, tanto al recordar datos como al realizar una acción."
          },
          "url": "https://thegraphweb.com/#hypothesis.evidence",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "hypothesis.currentQuestion",
          "predicate": "currentQuestion",
          "text": {
            "en": "The question this project tracks now is wider and harder to fake. Can an AI interface run for months with a Canonical Representation as its only source of knowledge? That means no second store of knowledge beside it.",
            "es": "La pregunta que sigue ahora el proyecto es más amplia y más difícil de amañar. ¿Puede una interfaz de IA funcionar durante meses con una Representación Canónica como única fuente de conocimiento? Es decir, sin ningún otro almacén de conocimiento al lado."
          },
          "url": "https://thegraphweb.com/#hypothesis.currentQuestion",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "vision",
      "kind": "Vision",
      "url": "https://thegraphweb.com/#vision",
      "relationships": [
        {
          "kind": "dependsOn",
          "target": "hypothesis"
        }
      ],
      "claims": [
        {
          "id": "vision.definition",
          "predicate": "definition",
          "text": {
            "en": "If a graph proves to work better, digital systems can be written for AI agents first. Today they are written for people, and agents have to guess the rest.",
            "es": "Si se demuestra que un grafo funciona mejor, los sistemas digitales podrán escribirse primero para agentes de IA. Hoy se escriben para personas, y los agentes tienen que deducir el resto."
          },
          "url": "https://thegraphweb.com/#vision.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "vision.status",
          "predicate": "status",
          "text": {
            "en": "Since it went live on 28 July 2026, AISwise has answered only from this project's own Canonical Representation. That is a live test of the idea, still small.",
            "es": "Desde que se publicó el 28 de julio de 2026, AISwise ha respondido solo a partir de la Representación Canónica de este proyecto. Es una prueba real de la idea, aún pequeña."
          },
          "url": "https://thegraphweb.com/#vision.status",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "iban-rios",
      "kind": "Person",
      "url": "https://thegraphweb.com/#iban-rios",
      "relationships": [
        {
          "kind": "builds",
          "target": "aiswise"
        }
      ],
      "claims": [
        {
          "id": "iban-rios.definition",
          "predicate": "definition",
          "text": {
            "en": "Iban Rios builds the systems behind sales and marketing work. Today he does AI GTM Engineering at Sesame HR. Before that, he spent seven years running paid media for global brands at Dentsu International and Omnicom/PHD. He also co-founded and led Chat Ergo Bot (2021 to 2023).",
            "es": "Iban Rios construye los sistemas detrás del trabajo de ventas y marketing. Hoy hace AI GTM Engineering en Sesame HR. Antes pasó siete años dirigiendo medios de pago para marcas globales en Dentsu International y Omnicom/PHD. También cofundó y dirigió Chat Ergo Bot (de 2021 a 2023)."
          },
          "url": "https://thegraphweb.com/#iban-rios.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "iban-rios.throughline",
          "predicate": "throughline",
          "text": {
            "en": "What links these roles: he treats a business task as a system to design, not as a set of campaigns to run by hand.",
            "es": "Lo que une estos puestos: trata una función de negocio como un sistema que diseñar, no como un conjunto de campañas que llevar a mano."
          },
          "url": "https://thegraphweb.com/#iban-rios.throughline",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "iban-rios.projects",
          "predicate": "projects",
          "text": {
            "en": "ibanrios.com is his own website. His other live projects are aiswise.com; validwise.com, which checks sites against EU rules; and geoadsfinder.com, which checks search results and spies on ads. In 2020 he made yometesteoencasa, Spain's first COVID-19 self-test tool.",
            "es": "ibanrios.com es su propia web. Sus otros proyectos en marcha son aiswise.com; validwise.com, que revisa sitios frente a las normas de la UE; y geoadsfinder.com, que revisa resultados de búsqueda y espía anuncios. En 2020 hizo yometesteoencasa, la primera herramienta de autotest de COVID-19 de España."
          },
          "url": "https://thegraphweb.com/#iban-rios.projects",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "aiswise",
      "kind": "Project",
      "url": "https://thegraphweb.com/#aiswise",
      "relationships": [
        {
          "kind": "consumes",
          "target": "the-graph-web"
        },
        {
          "kind": "waitlist",
          "target": "contact"
        }
      ],
      "claims": [
        {
          "id": "aiswise.availability",
          "predicate": "availability",
          "text": {
            "en": "You can use AISwise now at thegraphweb.com/chat. It is an AI chat that answers only from what this site has published, and turns down anything else.",
            "es": "Puedes usar AISwise ya en thegraphweb.com/chat. Es un chat de IA que responde solo a partir de lo que esta web ha publicado y rechaza todo lo demás."
          },
          "url": "https://thegraphweb.com/#aiswise.availability",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "aiswise.commercialStatus",
          "predicate": "commercialStatus",
          "text": {
            "en": "It is sold at aiswise.com.",
            "es": "Se vende en aiswise.com."
          },
          "url": "https://thegraphweb.com/#aiswise.commercialStatus",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "aiswise.waitlist",
          "predicate": "waitlist",
          "text": {
            "en": "To hear first if it is offered for other sites, join the waitlist through the contact form.",
            "es": "Para enterarte antes que nadie si se ofrece para otras webs, apúntate a la lista de espera desde el formulario de contacto."
          },
          "url": "https://thegraphweb.com/#aiswise.waitlist",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "architecture",
      "kind": "Architecture",
      "url": "https://thegraphweb.com/#architecture",
      "relationships": [
        {
          "kind": "implements",
          "target": "the-graph-web"
        }
      ],
      "claims": [
        {
          "id": "architecture.definition",
          "predicate": "definition",
          "text": {
            "en": "AISwise runs on Cloudflare and answers only from this Canonical Representation. It has no other source of knowledge and no general mode, so each reply either comes from it or says no.",
            "es": "AISwise funciona en Cloudflare y responde solo a partir de esta Representación Canónica. No tiene otra fuente de conocimiento ni un modo general, así que cada respuesta o sale de ella o dice que no."
          },
          "url": "https://thegraphweb.com/#architecture.definition",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "architecture.evidence",
          "predicate": "evidence",
          "text": {
            "en": "Each turn of each chat is logged to show which parts of it were used. That record, not a guess, decides when the Canonical Representation changes.",
            "es": "Cada turno de cada chat se registra para mostrar qué partes de ella se usaron. Ese registro, y no una suposición, decide cuándo cambia la Representación Canónica."
          },
          "url": "https://thegraphweb.com/#architecture.evidence",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "architecture.publication",
          "predicate": "publication",
          "text": {
            "en": "The Canonical Representation is a small file, written by hand. Each time the site is updated, it is published as JSON at thegraphweb.com/cr and as a plain web page at thegraphweb.com.",
            "es": "La Representación Canónica es un archivo pequeño, escrito a mano. Cada vez que se actualiza la web, se publica como JSON en thegraphweb.com/cr y como página web sencilla en thegraphweb.com."
          },
          "url": "https://thegraphweb.com/#architecture.publication",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "contact",
      "kind": "Action",
      "url": "https://thegraphweb.com/#contact",
      "relationships": [],
      "claims": [
        {
          "id": "contact.behaviour",
          "predicate": "behaviour",
          "text": {
            "en": "When you send a message, it is stored with the time, for a person to read. Nothing replies on its own.",
            "es": "Cuando envías un mensaje, se guarda con la hora para que una persona lo lea. Nada responde por sí solo."
          },
          "url": "https://thegraphweb.com/#contact.behaviour",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "contact.requiredFields",
          "predicate": "requiredFields",
          "text": {
            "en": "Name and email are always required.",
            "es": "El nombre y el email son siempre obligatorios."
          },
          "url": "https://thegraphweb.com/#contact.requiredFields",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "contact.waitlist",
          "predicate": "waitlist",
          "text": {
            "en": "Tick the waitlist box to hear first if AISwise is offered for other sites; then the message can be empty. The address joins the waitlist only after its owner opens the link sent to it and presses Confirm.",
            "es": "Marca la casilla de la lista de espera para enterarte antes que nadie si AISwise se ofrece para otras webs; entonces el mensaje puede quedar vacío. La dirección entra en la lista solo cuando su dueño abre el enlace que se le envía y pulsa el botón «Confirm»."
          },
          "url": "https://thegraphweb.com/#contact.waitlist",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "newsletter",
      "kind": "Action",
      "url": "https://thegraphweb.com/#newsletter",
      "relationships": [],
      "claims": [
        {
          "id": "newsletter.behaviour",
          "predicate": "behaviour",
          "text": {
            "en": "You give an email address, the only field, and it gets a link. It joins a list for news on how this project grows only after its owner opens that link and presses Confirm, within 48 hours.",
            "es": "Das una dirección de email, el único campo, y recibe un enlace. Entra en una lista de novedades sobre cómo avanza este proyecto solo cuando su dueño abre ese enlace y pulsa el botón «Confirm», en un plazo de 48 horas."
          },
          "url": "https://thegraphweb.com/#newsletter.behaviour",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "newsletter.sendStatus",
          "predicate": "sendStatus",
          "text": {
            "en": "Nothing has been sent to the list yet.",
            "es": "Todavía no se ha enviado nada a la lista."
          },
          "url": "https://thegraphweb.com/#newsletter.sendStatus",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "newsletter.unsubscribe",
          "predicate": "unsubscribe",
          "text": {
            "en": "That email also carries a link to leave the list at any time.",
            "es": "Ese email también lleva un enlace para salir de la lista cuando quieras."
          },
          "url": "https://thegraphweb.com/#newsletter.unsubscribe",
          "source": {
            "kind": "owner",
            "ref": "src/graph.ts at bffb901, approved there under the 2026-07-28 change-control rule"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "format-test",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#format-test",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "hypothesis"
        }
      ],
      "claims": [
        {
          "id": "format-test.definition",
          "predicate": "definition",
          "text": {
            "en": "The first test of the idea, run on 28 July 2026. It asked one thing: does a graph help an AI agent more than a well-written document with the same facts?",
            "es": "La primera prueba de la idea, hecha el 28 de julio de 2026. Preguntaba una sola cosa: ¿ayuda más un grafo a un agente de IA que un documento bien escrito con los mismos datos?"
          },
          "url": "https://thegraphweb.com/#format-test.definition",
          "validFrom": "2026-07-28",
          "source": {
            "kind": "doc",
            "ref": "experiments/m002-feasibility/decision-rule.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "format-test.method",
          "predicate": "method",
          "text": {
            "en": "The rule to read the result was written before any trial ran. The same made-up facts went out twice: once as a document, once as a graph. Each version got 4 questions on facts and 4 requests to take an action. Code scored the answers, not an AI judge.",
            "es": "La regla para leer el resultado se escribió antes de hacer ninguna prueba. Los mismos datos inventados se dieron dos veces: una como documento y otra como grafo. Cada versión recibió 4 preguntas sobre datos y 4 peticiones de hacer una acción. Las respuestas las puntuó código, no un juez de IA."
          },
          "url": "https://thegraphweb.com/#format-test.method",
          "source": {
            "kind": "doc",
            "ref": "experiments/m002-feasibility/decision-rule.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "format-test.evidence",
          "predicate": "evidence",
          "text": {
            "en": "Both versions scored 4 of 4 on facts and 4 of 4 on actions. The gap was zero, so the format alone gave no gain that could be measured.",
            "es": "Las dos versiones sacaron 4 de 4 en datos y 4 de 4 en acciones. La diferencia fue cero: el formato por sí solo no dio ninguna ventaja medible."
          },
          "url": "https://thegraphweb.com/#format-test.evidence",
          "source": {
            "kind": "measurement",
            "ref": "experiments/m002-feasibility/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "format-test.limitation",
          "predicate": "limitation",
          "text": {
            "en": "The test was easy on purpose: one action, four plain fields, and questions answered in one step. With every answer right in both versions, there was no room left to see a gap.",
            "es": "La prueba era fácil a propósito: una acción, cuatro campos sencillos y preguntas que se responden en un paso. Con todas las respuestas bien en las dos versiones, no quedaba margen para ver una diferencia."
          },
          "url": "https://thegraphweb.com/#format-test.limitation",
          "source": {
            "kind": "doc",
            "ref": "experiments/m002-feasibility/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "format-test.conclusion",
          "predicate": "conclusion",
          "text": {
            "en": "After it, the project stopped testing format alone. It now asks whether a real AI interface can run for months with a Canonical Representation as its only source.",
            "es": "Después de ella, el proyecto dejó de probar solo el formato. Ahora pregunta si una interfaz de IA real puede funcionar durante meses con una Representación Canónica como única fuente."
          },
          "url": "https://thegraphweb.com/#format-test.conclusion",
          "source": {
            "kind": "doc",
            "ref": "docs/sovereign-v2/DECISIONS.md, 2026-07-28"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "review-test",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#review-test",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "architecture"
        }
      ],
      "claims": [
        {
          "id": "review-test.definition",
          "predicate": "definition",
          "text": {
            "en": "A test of what it costs to write the knowledge, run on 30 July 2026. Can one AI write a first draft and another AI catch its mistakes?",
            "es": "Una prueba de lo que cuesta escribir el conocimiento, hecha el 30 de julio de 2026. ¿Puede una IA escribir un primer borrador y otra IA encontrar sus errores?"
          },
          "url": "https://thegraphweb.com/#review-test.definition",
          "validFrom": "2026-07-30",
          "source": {
            "kind": "doc",
            "ref": "experiments/m005-authoring-cost/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "review-test.method",
          "predicate": "method",
          "text": {
            "en": "Six short pieces of knowledge were written, with two mistakes put in on purpose. A reviewer that had not seen the answers checked them. In a second run, separate AI agents drafted, challenged and fixed the pieces in turn.",
            "es": "Se escribieron seis piezas cortas de conocimiento, con dos errores metidos a propósito. Las revisó un revisor que no había visto las respuestas. En una segunda ronda, varios agentes de IA distintos redactaron, cuestionaron y corrigieron las piezas por turnos."
          },
          "url": "https://thegraphweb.com/#review-test.method",
          "source": {
            "kind": "doc",
            "ref": "experiments/m006-pipeline/decision-rule.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "review-test.evidence",
          "predicate": "evidence",
          "text": {
            "en": "The reviewer caught both planted mistakes, and two more that no one had planted. In the second run, both mistakes were fixed and none got through. No right piece was challenged by mistake.",
            "es": "El revisor encontró los dos errores metidos y otros dos que nadie había metido. En la segunda ronda se corrigieron los dos errores y no pasó ninguno. Ninguna pieza correcta se cuestionó por error."
          },
          "url": "https://thegraphweb.com/#review-test.evidence",
          "source": {
            "kind": "measurement",
            "ref": "experiments/m005-authoring-cost/results.md; experiments/m006-pipeline/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "review-test.limitation",
          "predicate": "limitation",
          "text": {
            "en": "Six pieces is a small sample, and the mistakes were easy ones. The time a person would spend was never measured, so the cost question is still open.",
            "es": "Seis piezas son una muestra pequeña, y los errores eran fáciles. Nunca se midió el tiempo que le llevaría a una persona, así que la pregunta del coste sigue abierta."
          },
          "url": "https://thegraphweb.com/#review-test.limitation",
          "source": {
            "kind": "doc",
            "ref": "experiments/m005-authoring-cost/results.md; experiments/m006-pipeline/results.md"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "chat-log-review",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#chat-log-review",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "aiswise"
        }
      ],
      "claims": [
        {
          "id": "chat-log-review.definition",
          "predicate": "definition",
          "text": {
            "en": "A look at what people asked AISwise, done on 18 September 2026. It read 998 chat records, from 28 July to 6 September 2026.",
            "es": "Una revisión de lo que la gente preguntó a AISwise, hecha el 18 de septiembre de 2026. Leyó 998 registros del chat, del 28 de julio al 6 de septiembre de 2026."
          },
          "url": "https://thegraphweb.com/#chat-log-review.definition",
          "validFrom": "2026-09-18",
          "source": {
            "kind": "measurement",
            "ref": "experiments/m008-evidence-mining/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "chat-log-review.evidence",
          "predicate": "evidence",
          "text": {
            "en": "812 of 946 messages (86%) were greetings or tests with nothing to answer. Only 54 different messages asked something real, and 8.1% were about things the site does not cover. The most common question, «what is this?», already had an answer.",
            "es": "812 de 946 mensajes (el 86 %) eran saludos o pruebas sin nada que responder. Solo 54 mensajes distintos preguntaban algo real, y el 8,1 % trataba de cosas que el sitio no cubre. La pregunta más común, «¿qué es esto?», ya tenía respuesta."
          },
          "url": "https://thegraphweb.com/#chat-log-review.evidence",
          "source": {
            "kind": "measurement",
            "ref": "experiments/m008-evidence-mining/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "chat-log-review.conclusion",
          "predicate": "conclusion",
          "text": {
            "en": "So the knowledge was not what held the chat back, and it was not made bigger. One part was fixed instead: it still said the project had no page for people, which was no longer true.",
            "es": "Así que el conocimiento no era lo que frenaba al chat, y no se amplió. En cambio se corrigió una parte: aún decía que el proyecto no tenía página para personas, y eso ya no era cierto."
          },
          "url": "https://thegraphweb.com/#chat-log-review.conclusion",
          "source": {
            "kind": "doc",
            "ref": "experiments/m008-evidence-mining/results.md; docs/sovereign-v2/DECISIONS.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "chat-log-review.limitation",
          "predicate": "limitation",
          "text": {
            "en": "A later count found that most records came from the launch day, 28 July. So they were mostly tests, not real use.",
            "es": "Un recuento posterior vio que la mayoría de los registros eran del día del lanzamiento, el 28 de julio. Así que eran sobre todo pruebas, no uso real."
          },
          "url": "https://thegraphweb.com/#chat-log-review.limitation",
          "source": {
            "kind": "doc",
            "ref": "docs/sovereign-v2/DECISIONS.md (M012: 764 of 993 records are from 28 July)"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "plain-page-test",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#plain-page-test",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "hypothesis"
        }
      ],
      "claims": [
        {
          "id": "plain-page-test.definition",
          "predicate": "definition",
          "text": {
            "en": "A test of how well an AI agent reads this site, run on 23 September 2026. It compared the old front page, the JSON at thegraphweb.com/cr and a new plain web page with the same content.",
            "es": "Una prueba de lo bien que un agente de IA lee este sitio, hecha el 23 de septiembre de 2026. Comparó la portada antigua, el JSON de thegraphweb.com/cr y una página web sencilla nueva con el mismo contenido."
          },
          "url": "https://thegraphweb.com/#plain-page-test.definition",
          "validFrom": "2026-09-23",
          "source": {
            "kind": "doc",
            "ref": "experiments/m010-base-page/decision-rule.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "plain-page-test.method",
          "predicate": "method",
          "text": {
            "en": "The rule was written before the new page existed. An AI agent read each of the three pages and answered 6 questions, twice: 36 reads in all. A grader scored the answers without knowing which page each one came from.",
            "es": "La regla se escribió antes de que existiera la página nueva. Un agente de IA leyó cada una de las tres páginas y respondió 6 preguntas, dos veces: 36 lecturas en total. Un evaluador puntuó las respuestas sin saber de qué página venía cada una."
          },
          "url": "https://thegraphweb.com/#plain-page-test.method",
          "source": {
            "kind": "doc",
            "ref": "experiments/m010-base-page/decision-rule.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "plain-page-test.evidence",
          "predicate": "evidence",
          "text": {
            "en": "The old front page got 2 of 12 right, the JSON 11 of 12 and the plain page 12 of 12, with no made-up answers. The plain page lost nothing against the JSON.",
            "es": "La portada antigua acertó 2 de 12, el JSON 11 de 12 y la página sencilla 12 de 12, sin respuestas inventadas. La página sencilla no perdió nada frente al JSON."
          },
          "url": "https://thegraphweb.com/#plain-page-test.evidence",
          "source": {
            "kind": "measurement",
            "ref": "experiments/m010-base-page/results.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "plain-page-test.conclusion",
          "predicate": "conclusion",
          "text": {
            "en": "So the plain page became the front page, and the chat moved to thegraphweb.com/chat. The old page lacked content; its format was not the problem.",
            "es": "Así que la página sencilla pasó a ser la portada, y el chat se movió a thegraphweb.com/chat. A la página antigua le faltaba contenido; el problema no era su formato."
          },
          "url": "https://thegraphweb.com/#plain-page-test.conclusion",
          "source": {
            "kind": "doc",
            "ref": "docs/sovereign-v2/DECISIONS.md, 2026-09-23"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "plain-page-test.limitation",
          "predicate": "limitation",
          "text": {
            "en": "One reader and two tries per question are a small sample: 12 against 11 could be chance. And no page told an agent how to take an action yet.",
            "es": "Un solo lector y dos intentos por pregunta son una muestra pequeña: 12 frente a 11 puede ser azar. Y ninguna página explicaba aún a un agente cómo hacer una acción."
          },
          "url": "https://thegraphweb.com/#plain-page-test.limitation",
          "source": {
            "kind": "doc",
            "ref": "experiments/m010-base-page/results.md"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "idea-review",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#idea-review",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "hypothesis"
        },
        {
          "kind": "tests",
          "target": "vision"
        }
      ],
      "claims": [
        {
          "id": "idea-review.definition",
          "predicate": "definition",
          "text": {
            "en": "A review of what this project claims, done on 17 October 2026. It checked each claim against the project's own code, data and tests, and against outside studies.",
            "es": "Una revisión de lo que afirma este proyecto, hecha el 17 de octubre de 2026. Contrastó cada afirmación con el código, los datos y las pruebas del propio proyecto, y con estudios de fuera."
          },
          "url": "https://thegraphweb.com/#idea-review.definition",
          "validFrom": "2026-10-17",
          "source": {
            "kind": "doc",
            "ref": "sovereign/audits/AUDIT-2026-10-17-thegraphweb-tesis.html"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "idea-review.evidence",
          "predicate": "evidence",
          "text": {
            "en": "What held up: AI models do use websites as a source of knowledge. Agents that take actions on a site are real, but early.",
            "es": "Lo que se sostiene: los modelos de IA sí usan las webs como fuente de conocimiento. Los agentes que hacen acciones en un sitio son reales, pero aún tempranos."
          },
          "url": "https://thegraphweb.com/#idea-review.evidence",
          "source": {
            "kind": "doc",
            "ref": "sovereign/audits/AUDIT-2026-10-17-thegraphweb-tesis.html"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "idea-review.evidence-2",
          "predicate": "evidence",
          "text": {
            "en": "What did not: no test here found that a special format helps an agent by itself. Two tests gave the same, or nearly the same, score in each format: 8 of 8 against 8 of 8, and 12 of 12 against 11 of 12.",
            "es": "Lo que no: ninguna prueba de aquí encontró que un formato especial ayude por sí solo a un agente. Dos pruebas dieron la misma puntuación, o casi, en cada formato: 8 de 8 frente a 8 de 8, y 12 de 12 frente a 11 de 12."
          },
          "url": "https://thegraphweb.com/#idea-review.evidence-2",
          "source": {
            "kind": "doc",
            "ref": "sovereign/audits/AUDIT-2026-10-17-thegraphweb-tesis.html"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "idea-review.evidence-3",
          "predicate": "evidence",
          "text": {
            "en": "Nor is it true today that people no longer browse the web. In 2025, AI bots made about 4.2% of the requests for web pages that Cloudflare saw.",
            "es": "Tampoco es cierto hoy que la gente ya no navegue por la web. En 2025, los bots de IA hicieron alrededor del 4,2 % de las peticiones de páginas web que vio Cloudflare."
          },
          "url": "https://thegraphweb.com/#idea-review.evidence-3",
          "source": {
            "kind": "doc",
            "ref": "sovereign/audits/AUDIT-2026-10-17-thegraphweb-tesis.html (Cloudflare, 2025)"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "live-chat-tests",
      "kind": "Experiment",
      "url": "https://thegraphweb.com/#live-chat-tests",
      "relationships": [
        {
          "kind": "partOf",
          "target": "the-graph-web"
        },
        {
          "kind": "tests",
          "target": "aiswise"
        }
      ],
      "claims": [
        {
          "id": "live-chat-tests.definition",
          "predicate": "definition",
          "text": {
            "en": "Real attacks and vague questions sent to the live chat after each change to it, from 15 to 18 October 2026.",
            "es": "Ataques reales y preguntas vagas enviados al chat en vivo después de cada cambio, del 15 al 18 de octubre de 2026."
          },
          "url": "https://thegraphweb.com/#live-chat-tests.definition",
          "validFrom": "2026-10-15",
          "source": {
            "kind": "doc",
            "ref": "docs/security.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "live-chat-tests.method",
          "predicate": "method",
          "text": {
            "en": "16 attacks try to make the chat leak its rules, link to other sites, invent contacts or act without asking. A person also reads each answer, since the automatic checks can miss things.",
            "es": "16 ataques intentan que el chat revele sus reglas, enlace a otros sitios, invente contactos o actúe sin preguntar. Una persona lee además cada respuesta, porque las comprobaciones automáticas pueden pasar cosas por alto."
          },
          "url": "https://thegraphweb.com/#live-chat-tests.method",
          "source": {
            "kind": "doc",
            "ref": "docs/security.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "live-chat-tests.evidence",
          "predicate": "evidence",
          "text": {
            "en": "Each of the six runs passed 16 of 16. In the first run, though, a person reading the answers saw the chat list its rules in its own words. That was closed, and the case is now one of the tests.",
            "es": "Cada una de las seis rondas pasó 16 de 16. Pero en la primera, una persona que leía las respuestas vio que el chat contaba sus reglas con sus propias palabras. Eso se cerró, y el caso es ahora una de las pruebas."
          },
          "url": "https://thegraphweb.com/#live-chat-tests.evidence",
          "source": {
            "kind": "measurement",
            "ref": "docs/security.md (runs of 2026-10-15 to 2026-10-18)"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "live-chat-tests.evidence-2",
          "predicate": "evidence",
          "text": {
            "en": "Of 12 vague questions, 1 got an answer that copied the site's own words, «these Nodes». After a fix, none did.",
            "es": "De 12 preguntas vagas, 1 recibió una respuesta que copiaba las palabras del propio sitio, «these Nodes». Tras un arreglo, ninguna."
          },
          "url": "https://thegraphweb.com/#live-chat-tests.evidence-2",
          "source": {
            "kind": "measurement",
            "ref": "docs/sovereign-v2/DECISIONS.md, 2026-10-17; docs/security.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "live-chat-tests.limitation",
          "predicate": "limitation",
          "text": {
            "en": "Passing shows that these attacks fail. It does not prove that no other attack works.",
            "es": "Pasarlas muestra que estos ataques fallan. No demuestra que no funcione ningún otro."
          },
          "url": "https://thegraphweb.com/#live-chat-tests.limitation",
          "source": {
            "kind": "doc",
            "ref": "docs/security.md"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "ibanrios-com",
      "kind": "Project",
      "url": "https://thegraphweb.com/#ibanrios-com",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        }
      ],
      "claims": [
        {
          "id": "ibanrios-com.definition",
          "predicate": "definition",
          "text": {
            "en": "ibanrios.com is Iban's own website, in English and Spanish.",
            "es": "ibanrios.com es la web propia de Iban, en inglés y en español."
          },
          "url": "https://thegraphweb.com/#ibanrios-com.definition",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios AGENTS.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "ibanrios-com.commercialStatus",
          "predicate": "commercialStatus",
          "text": {
            "en": "There he builds AI agents, automations and software that hand over finished work. You pay for each result you accept.",
            "es": "Allí construye agentes de IA, automatizaciones y software que entregan trabajo terminado. Se paga por cada resultado que se acepta."
          },
          "url": "https://thegraphweb.com/#ibanrios-com.commercialStatus",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios src/data/home/en.json"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "validwise",
      "kind": "Project",
      "url": "https://thegraphweb.com/#validwise",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        }
      ],
      "claims": [
        {
          "id": "validwise.definition",
          "predicate": "definition",
          "text": {
            "en": "ValidWise, at validwise.com, checks websites in Europe against the rules they must meet.",
            "es": "ValidWise, en validwise.com, revisa webs de Europa frente a las normas que deben cumplir."
          },
          "url": "https://thegraphweb.com/#validwise.definition",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/validwise README.md"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "validwise.detail",
          "predicate": "detail",
          "text": {
            "en": "It runs 70 checks on its own, in five groups. Accessibility has 30, data protection 15, security headers 14, the AI Act 5 and other legal needs 6.",
            "es": "Hace 70 comprobaciones automáticas, en cinco grupos. Accesibilidad tiene 30, protección de datos 15, cabeceras de seguridad 14, la Ley de IA 5 y otros requisitos legales 6."
          },
          "url": "https://thegraphweb.com/#validwise.detail",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios src/data/why-me/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "validwise.principle",
          "predicate": "principle",
          "text": {
            "en": "It finds problems. It does not certify that a site complies, and its results are not legal advice.",
            "es": "Encuentra problemas. No certifica que un sitio cumpla, y sus resultados no son asesoría legal."
          },
          "url": "https://thegraphweb.com/#validwise.principle",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/validwise web/src/i18n/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "validwise.availability",
          "predicate": "availability",
          "text": {
            "en": "Single scans are free. A Pro plan is on request.",
            "es": "Los análisis sueltos son gratis. El plan Pro se pide."
          },
          "url": "https://thegraphweb.com/#validwise.availability",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/validwise web/src/i18n/en.json"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "geoadsfinder",
      "kind": "Project",
      "url": "https://thegraphweb.com/#geoadsfinder",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        }
      ],
      "claims": [
        {
          "id": "geoadsfinder.definition",
          "predicate": "definition",
          "text": {
            "en": "GeoAdsFinder, at geoadsfinder.com, is a tool for people who work in search and paid ads.",
            "es": "GeoAdsFinder, en geoadsfinder.com, es una herramienta para quien trabaja en buscadores y anuncios de pago."
          },
          "url": "https://thegraphweb.com/#geoadsfinder.definition",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/geoadsfinder web/src/i18n/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "geoadsfinder.detail",
          "predicate": "detail",
          "text": {
            "en": "It shows Google results as seen from any country or city. It also shows the ads that rivals run on Meta, Google, TikTok and LinkedIn.",
            "es": "Muestra los resultados de Google como se ven desde cualquier país o ciudad. También muestra los anuncios que la competencia publica en Meta, Google, TikTok y LinkedIn."
          },
          "url": "https://thegraphweb.com/#geoadsfinder.detail",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/geoadsfinder web/src/i18n/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "geoadsfinder.motivation",
          "predicate": "motivation",
          "text": {
            "en": "It began at Dentsu International, for a problem the whole network had. Then the whole network used it.",
            "es": "Nació en Dentsu International, para un problema que tenía toda la red. Luego la usó toda la red."
          },
          "url": "https://thegraphweb.com/#geoadsfinder.motivation",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios src/data/why-me/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "geoadsfinder.availability",
          "predicate": "availability",
          "text": {
            "en": "The basic version is free and needs no account. A Pro version is on request.",
            "es": "La versión básica es gratis y no pide cuenta. La versión Pro se pide."
          },
          "url": "https://thegraphweb.com/#geoadsfinder.availability",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/geoadsfinder README.md"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "yometesteoencasa",
      "kind": "Project",
      "url": "https://thegraphweb.com/#yometesteoencasa",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        },
        {
          "kind": "ledTo",
          "target": "chat-ergo-bot"
        }
      ],
      "claims": [
        {
          "id": "yometesteoencasa.definition",
          "predicate": "definition",
          "text": {
            "en": "#YoMeTesteoEnCasa was an online COVID-19 self-test for people in Spain. It went live on 14 March 2020.",
            "es": "#YoMeTesteoEnCasa fue un test online de COVID-19 para la gente en España. Se publicó el 14 de marzo de 2020."
          },
          "url": "https://thegraphweb.com/#yometesteoencasa.definition",
          "validFrom": "2020-03-14",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/yometesteoencasa-hostinger index.php (datePublished)"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "yometesteoencasa.motivation",
          "predicate": "motivation",
          "text": {
            "en": "Its only aim was to take pressure off the emergency services during the first wave.",
            "es": "Su único objetivo era descongestionar los servicios de emergencia durante la primera ola."
          },
          "url": "https://thegraphweb.com/#yometesteoencasa.motivation",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/yometesteoencasa-hostinger index.php"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "yometesteoencasa.detail",
          "predicate": "detail",
          "text": {
            "en": "Iban built it in 24 hours with two friends. Expansión and Xataka, among others, wrote about it.",
            "es": "Iban lo hizo en 24 horas con dos amigos. Expansión y Xataka, entre otros, escribieron sobre él."
          },
          "url": "https://thegraphweb.com/#yometesteoencasa.detail",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios src/data/why-me/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "yometesteoencasa.status",
          "predicate": "status",
          "text": {
            "en": "It was active in March and April 2020, and it is no longer online.",
            "es": "Estuvo activo en marzo y abril de 2020, y ya no está en línea."
          },
          "url": "https://thegraphweb.com/#yometesteoencasa.status",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/yometesteoencasa-hostinger index.php; RESCATE.md (domain parked)"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    },
    {
      "id": "chat-ergo-bot",
      "kind": "Project",
      "url": "https://thegraphweb.com/#chat-ergo-bot",
      "relationships": [
        {
          "kind": "authoredBy",
          "target": "iban-rios"
        },
        {
          "kind": "ledTo",
          "target": "aiswise"
        }
      ],
      "claims": [
        {
          "id": "chat-ergo-bot.definition",
          "predicate": "definition",
          "text": {
            "en": "Chat Ergo Bot was a platform to build chatbots with conversational AI, from before ChatGPT. With no code, it turned a website into a chatbot that could also talk by voice.",
            "es": "Chat Ergo Bot fue una plataforma para crear chatbots con IA conversacional, de antes de ChatGPT. Sin programar, convertía una web en un chatbot que también hablaba por voz."
          },
          "url": "https://thegraphweb.com/#chat-ergo-bot.definition",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/chatergobot llms.txt; IbanRios/ibanrios src/data/why-me/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "chat-ergo-bot.detail",
          "predicate": "detail",
          "text": {
            "en": "Lanzadera picked it in 2023. In 2022 El Referente put it in the top 75 startups of Spain and Portugal, and it was a Planeta Chatbot finalist.",
            "es": "Lanzadera la seleccionó en 2023. En 2022 El Referente la puso entre las 75 mejores startups de España y Portugal, y fue finalista de Planeta Chatbot."
          },
          "url": "https://thegraphweb.com/#chat-ergo-bot.detail",
          "source": {
            "kind": "doc",
            "ref": "IbanRios/ibanrios src/data/why-me/en.json"
          },
          "approvedAt": "2026-10-18"
        },
        {
          "id": "chat-ergo-bot.status",
          "predicate": "status",
          "text": {
            "en": "Iban co-founded and led it from 2021, and it closed in 2023. AISwise grew out of it.",
            "es": "Iban la cofundó y la dirigió desde 2021, y cerró en 2023. De ella nació AISwise."
          },
          "url": "https://thegraphweb.com/#chat-ergo-bot.status",
          "source": {
            "kind": "owner",
            "ref": "Iban, 2026-10-18: «fue cofounded pero liderado por mí»; foundingDate in IbanRios/chatergobot index.html"
          },
          "approvedAt": "2026-10-18"
        }
      ]
    }
  ]
}