# « What if it makes things up? » — getting an agent to cite its sources > A model produces a plausible sentence, not a true one. The four designs that make invention visible, and the single rule that matters: every claim points to its source. **Version HTML :** https://agence-intelligence-artificielle.eu/en/blog/what-if-it-makes-things-up-getting-an-agent-to-cite **Langue :** en-GB --- **On this page** The single rule: cite Four designs that work Two false fixes How we check that it holds Responsible AI # « What if it makes things up? » — getting an agent to cite its sources It is the first question in every workshop, and it is the right one. A language model produces a plausible sentence; nothing in its mechanics guarantees it is true. The craft is in making the error visible. 14 August 20266 min readBy Sébastien Joumel **You do not remove invention, you make it visible**No design eliminates the risk. A good design makes a mistake obvious instead of letting it pass unnoticed. ## The single rule: cite Every claim an agent produces points to the document and version it came from. Not « according to our procedures » — the exact reference, clickable, checkable in three seconds. The rule looks modest. It changes everything: it turns review from an act of trust into an act of verification. Someone who must believe reviews badly; someone who can check reviews fast. > If a sentence has no source, it has no business in the output. ## Four designs that work ### Answer from the documents, not from memory The agent searches your base first, then writes from what it found. If it finds nothing, it must say so — not fill the gap with what it « knows ». ### Allow « I don’t know » An agent that is not allowed to fail will invent. The circuit must provide a « not in the base » exit that escalates to a person — and that exit should be frequent at the start. ### Separate extraction from writing Extracting a date from a document is verifiable. Writing a paragraph is not. When both are mixed, nobody knows what they are reviewing. ### Log the inputs The log keeps what the agent read, not only what it produced. That is what lets you fix the cause instead of the output. ## Two false fixes **« We’ll use a more powerful model. »** A better model reduces how often errors happen, not their nature. On a question whose answer is nowhere in your documents, the best model on the market will also invent — just more convincingly. **« We’ll review everything. »** Nobody reviews three hundred outputs a day. After two weeks, review becomes a glance. Better to review ten per cent properly than one hundred per cent in passing. ## How we check that it holds When a system opens, we build a set of real cases — forty to a hundred — whose correct answer is known. We measure three things: the share of correct answers, the share of justified « don’t knows », and the share of silent errors. Only the third really matters. That case set is not only for go-live: we replay it on every model or prompt change. It is what lets you say, with numbers, whether an update improved or degraded the system. **Read on the site** - [AI for banking](https://agence-intelligence-artificielle.eu/en/ai-banking) — answer by quoting the file, never from memory - [AI for legal teams](https://agence-intelligence-artificielle.eu/en/ai-legal) — the finding of a departure, not the advice - [AI for healthcare](https://agence-intelligence-artificielle.eu/en/ai-healthcare) — cite the version of a procedure, nothing else - [The survey of your ten systems](https://agence-intelligence-artificielle.eu/en/contact) — free, delivered within 72 hours ## Other notes [Systems · 2 September 2026Your document base, or why the agent answers beside the pointAn agent is only worth what it can read. The four defects that produce answers beside the point — duplicates, undated versions, scanned PDFs, implicit context — and the preparation that fixes them.](https://agence-intelligence-artificielle.eu/en/blog/why-your-agent-answers-beside-the-point)[Responsible AI · 8 September 2026The EU AI Act: what it means if you sell into EuropeApplication dates for Regulation (EU) 2024/1689, what it asks of a company that uses AI without building it, and why it reaches UK businesses that serve EU customers.](https://agence-intelligence-artificielle.eu/en/blog/eu-ai-act-what-it-means-if-you-sell-into-europe)[Systems · 10 July 2026Getting cited by ChatGPT and Perplexity: what actually worksGenerative engines cite what they can read, structure and verify. Five concrete moves — including serving your pages as Markdown — and what is a waste of time.](https://agence-intelligence-artificielle.eu/en/blog/getting-cited-by-chatgpt-and-perplexity-what-works) [All notes →](https://agence-intelligence-artificielle.eu/en/blog) First call — fifteen minutes ## We will tell you which system to open first. Describe the task that costs you the most. We will tell you what is feasible, how long it takes and what it costs. If the answer is no, you will leave with the two reasons why. [Request the survey](https://agence-intelligence-artificielle.eu/en/contact) [Write to the agency](https://agence-intelligence-artificielle.eu/en/contact) --- ## Autres langues - fr-FR : https://agence-intelligence-artificielle.eu/blog/et-sil-invente-faire-citer-ses-sources-a-un-agent.md - en-GB : https://agence-intelligence-artificielle.eu/en/blog/what-if-it-makes-things-up-getting-an-agent-to-cite.md - de-DE : https://agence-intelligence-artificielle.eu/de/blog/und-wenn-er-sich-etwas-ausdenkt-quellen-nennen.md - es-ES : https://agence-intelligence-artificielle.eu/es/blog/y-si-se-lo-inventa-hacer-que-un-agente-cite-sus-fuentes.md - it-IT : https://agence-intelligence-artificielle.eu/it/blog/e-se-si-inventa-far-citare-le-fonti-a-un-agente.md - pt-PT : https://agence-intelligence-artificielle.eu/pt/blog/e-se-inventar-fazer-um-agente-citar-as-fontes.md