- [Home](https://www.agenticsearch.at/en/) - [Services](https://www.agenticsearch.at/en/services/) - AI Brain # AI Brain, the fact base for AI systems ## What machines know about you before they answer. Before a system writes an answer about your company, it gathers facts. The AI Brain is the ordered version of those facts: unambiguous, maintained in one place, machine readable and dated. [Technical background](https://www.agenticsearch.at/en/knowledge/schema-markup-for-ai/) [Discuss your fact base](https://www.agenticsearch.at/en/contact/)  Definition ## What an AI Brain is. Verifiable on this site We do not only build this, we run it. You can open all three of these right now and look at them. - [llms.txt](https://www.agenticsearch.at/llms.txt) live - [robots.txt](https://www.agenticsearch.at/robots.txt) live - JSON-LD graph per page in the source - Markdown version per page address plus index.md The facts section of this site is being built and gets linked here once it stands. An AI Brain is a company's structured fact base, built for machine reading. It answers the questions a system asks before speaking about you: who exactly is this, what does the company do, since when, where, for whom, and which details are evidenced. The term is ours, the components are not. It covers entities, structured data, consistent details across every source, a curated short version of the site for language models, and a process that keeps all of it current. Two things separate it from ordinary schema work. First, everything hangs in one connected graph with stable anchors rather than sitting as fragments across pages. Second, every fact carries a source and a review date. Without that a fact base goes stale quietly and nobody notices. Components ## What it is made of. ### Entity model Company, people, products and locations as clearly bounded things with unambiguous names and defined relationships. ### Fact pages One page per entity with evidenced details, a source note and a visible review date. Facts to look up rather than marketing copy. ### Connected graph JSON-LD with stable anchors that any page can reference. One graph instead of scattered fragments. ### Short version for models An llms.txt that orders the site, plus a Markdown version per page. Agents prefer the source over parsing HTML. ### External alignment The same details in directories, profiles and press pages. Contradictions get hunted down rather than only fixed at home. ### Maintenance process A fixed review interval, named ownership, documented changes. The part almost everyone leaves out. Boundaries ## What the AI Brain does not cover. A clean fact base makes systems describe you correctly when they talk about you. It does not make them talk about you. The mention itself depends on visibility and sources, which is what [GEO](https://www.agenticsearch.at/en/services/geo-agency/) and [LLMO](https://www.agenticsearch.at/en/services/llmo-agency/) work on. The reverse is also true: the best visibility does little when what spreads is wrong. So the AI Brain sits near the start of almost every project, right after the first measurement. Common questions ## About the AI Brain. **What is an AI Brain?** The structured fact base about your company, built for machine reading: entities, evidenced details, one connected JSON-LD graph, a short version for language models, and a process keeping it current. **Is this just another word for structured data?** Structured data is one part of it. The difference lies in the connection and the upkeep: everything hangs in a graph with stable anchors, every fact carries a source and a review date, and somebody is named as responsible. **Which facts belong in it?** Anything somebody might ask about that can be evidenced: legal name, founding, locations, services, audiences, people with roles, references, awards, memberships. Nothing unverifiable, and no figures without a source. **How is this different from an internal knowledge base?** An internal knowledge base serves employees and sits behind a login. The AI Brain is public and built for machines. The two can share content while serving different purposes. **How often does it need maintaining?** We work with a review every three to six months, plus ad hoc updates for changes like a new location or a change of management. The review date sits visibly on the page so a long gap becomes obvious. **What do grounding pages have to do with it?** Grounding pages are the visible form of the fact base: one page per entity with evidenced details instead of marketing copy. They are the part of the AI Brain humans can read too. **Can we run this ourselves?** Yes, and many do after the build. We create the structure, set up the maintenance process and hand over. Keeping the upkeep with us is optional, not required. **What happens when facts change?** A change is made in one place and flows from there into every output: visible page, structured data, short version for models. Then comes the external alignment across directories and profiles. That second part is the one usually forgotten. **Do we need a new CMS for this?** Usually not. What matters is being able to generate structured data from the same source as the visible text. Where that is impossible we say so up front, rather than maintaining two copies that drift apart. **How do we measure whether the AI Brain works?** By two things: whether details about you are reproduced correctly in AI answers, and whether systems cite your own pages rather than third-party directories. Both run through the same question set as the other measurements. **What about confidential details?** They do not belong in it. The AI Brain only holds what is already public or meant to be. Revenue, unapproved client lists and internal metrics stay out. Difference ## How others do this, and how we do. Not a judgement on individual agencies, but a description of two approaches. The right column is checkable, because this website itself is built that way. Difference Usual agency approach agenticsearch.at Starting point A proposal after a first call A measurement with a date, before anything gets recommended Basis Keywords from classic search A question set of real prompts, clustered and prioritised Proof of success Rankings and a visibility index Mentions per question and system, with the raw data Reporting A monthly dashboard with curves The same measurement repeated, change called only after three runs When unsure Try it and see what happens We say what is not evidenced, and write it on the page Tooling Agency licences, access ends with the contract The question set and the data are yours, afterwards too Scope As many channels as possible from one supplier A capped number of concurrent projects, otherwise we turn it down Your guides ## You play the lead. You know your business, your customers and your offer better than any agency does. What is usually missing is solid numbers on how AI systems talk about it, and an order in which to tackle that. [More about us](https://www.agenticsearch.at/en/about/) [Request a call](https://www.agenticsearch.at/en/contact/) Who is behind this two people - Florian Hieß measurement, technology, structure - Romana Hieß day to day support and implementation - Operated by Digital Wings GmbH, founded 2010 - Client projects 100+ The number of concurrent projects is capped. When a piece of work needs more capacity than exists, we turn it down rather than take it on. 01 ### You know where you stand It starts with a measurement, not a proposal. You see in black and white which AI answers name your company and which do not, with a date and a question set. 02 ### You know what comes next Instead of a list of everything, you get an order. Every step has a result you can verify yourself, including without us. 03 ### You keep the decisions Your offer and your positioning stay yours. We tell you what stands in the way of visibility and take on the technical part of it. Next step ## Let us find out where you stand in AI answers. One conversation, 30 minutes, no sales pressure. You leave with an honest read on whether this is worth the effort for you. [Look at the audit first](https://www.agenticsearch.at/en/services/ai-visibility-audit/) [Request a call](https://www.agenticsearch.at/en/contact/)