--- title: "Business history and education videos with sources" description: "Tell business history and educational stories with checked dates, names and figures, and with care around real people. This is how CLAQON works." url: https://claqon.com/en/use-cases/history-and-education/ language: en updated: 2026-10-12 alternates: es: https://claqon.com/casos-de-uso/historia-y-divulgacion/ pt: https://claqon.com/pt/casos-de-uso/historia-e-divulgacao/ --- Use case · History and education # Business history and education, with dates and figures that can be checked. Business history and educational channels live on telling stories well and getting them right. CLAQON builds each piece on verbatim excerpts from its sources, separates facts from interpretation and warns you when a generated scene could show a real person. [Pre-register](https://claqon.com/en/pre-registration/#configurador) [See each piece’s compliance record](https://claqon.com/en/features/compliance-record/) ![Account editor on the Content tab: channel niche, video type, editorial focus, format and duration; on the left, the accounts with language and networks.](https://claqon.com/recursos/img/capturas/en/cuentas-perfil-1600.webp) Real CLAQON interface with sample data. The challenge ## One wrong date is enough to lose trust In a company’s history or an inventor’s biography, mistakes stand out: a date that does not add up, a figure rounded up, a misattributed quote or an anachronism in the image. AI tends to fill gaps with whatever sounds plausible. These topics are also full of real people. A generated scene showing the face of a business leader or a historical figure can raise image rights and impersonation issues, and generic stock footage is not footage of the event you are describing. How CLAQON solves it ## Rigour in the script, caution in the image - **Video types designed for storytelling**: “Narrated story” follows a chronological arc with context and resolution, and separates documented facts from interpretation. “Case study” covers the problem, the decision, the outcome and the lesson, and distinguishes correlation from causation. - **Dates, names and figures with a source**: The script can only cite excerpts that appear verbatim in the sources. HARD QC rejects figures that are not in the evidence and unsupported absolutes and superlatives, before anything is spent on voice. [Source verification](https://claqon.com/en/features/source-verification/). - **Real-people guard**: It detects proper names in the narration of generated scenes. By default, it warns you. If you choose to divert, the scene switches to stock clips or, if that is not possible, it is generated with the instruction not to show the person photorealistically: an illustration, objects or places. It is a heuristic, not a foolproof filter. - **Stock clips by default**: Without enabling anything paid, scenes use licensed stock clips, and the instructions ask for them not to be presented as real footage of the event. AI image and video are optional. [Voice and visuals](https://claqon.com/en/features/voice-and-visuals/). - **Disclosure and provenance**: If any scene is a generated image or video, the piece is disclosed as synthetic content on YouTube. Each render leaves a provenance record with the music, the clips, the providers and their licences. [Compliance record](https://claqon.com/en/features/compliance-record/). - **Long format Beta**: History often needs more time: in addition to the vertical short of 30 to 180 seconds, CLAQON produces horizontal long videos of 4 to 12 minutes for YouTube. A typical workflow ## From a historical question to a published piece An example of the process. The stages are the real ones; the topic, the format and what you enable depend on your account. 1. **A specific topic in the queue** For example, the decision that changed the course of a well-known company. The more specific the question, the easier it is to back it up with sources. 2. **Research and verbatim excerpts** CLAQON gathers the sources, or uses yours, and discards any excerpt that does not appear word for word or that is off topic. 3. **Narrated script and check** The AI model writes a narrated story or a case study. If a date or a figure is not in the evidence, HARD QC asks for a repair or holds the script for review. 4. **Scenes, with caution** The narration sets the pace of the scenes. If you have enabled AI images and a scene names a person, the guard warns you or diverts the scene, as you choose. 5. **Review and publishing** For delicate pieces, produce in draft mode, review the edit and publish it by hand from the storyboard. The cost and the provenance are recorded. What you control ## You decide how much risk you take in each scene - The account’s niche: “History and biographies”, “Finance and business” or a custom one. - Video type, format and length of each piece. - Real-people guard mode: warn or divert. - Visual profile: free stock clips or AI image and video, with its own budget. - The account’s editorial approach, in up to 1,200 characters. - Spend caps per video, per day and per month. Limits ## What you should know before you start - The real-people guard is a heuristic: it looks for proper names of two or more words and can get it wrong in both directions. - It does not generate maps, reconstructions or period portraits: you provide the real archive images, for example in the Editing Studio (beta). - It does not carry out a legal review: image rights and editorial responsibility are yours. - Topics with scarce or contradictory sources may be stopped at the check instead of being published. - Long-form video is in beta. Frequently asked questions ## Common questions from history channels ### Can CLAQON show the face of a historical figure or a real business leader? By default, scenes use generic stock clips, which are not images of that person. If you enable AI images, the real-people guard warns you when the narration names someone and, if you choose to divert, it switches the scene to stock clips or asks for a non-photorealistic illustration. Deciding what to show is still your responsibility. ### What happens if a date or a figure does not appear in the sources? HARD QC rejects it. CLAQON attempts a repair that fixes only what was flagged, without making up quotes, and if that is not enough it holds the script for review before spending on voice. ### Can I use my own articles or documents as a source? Yes, if they are published at a URL. In a production launched from the panel you paste the addresses, CLAQON fetches their content through the safe network and the script cites your links without going through the paid search. ### Does it work for long videos? Yes, in beta: it produces horizontal videos of 4 to 12 minutes for YouTube. The vertical short, of 30 to 180 seconds, is the most tested format. ### How is the use of AI disclosed? When any scene is a generated image or video, CLAQON marks the piece as synthetic content when it uploads it to YouTube. Synthetic voice on its own does not trigger that disclosure. [Guide: how to label AI-made content](https://claqon.com/en/resources/labeling-ai-content/). Early access ## Tell history with sources and without surprises Tell us which history or educational channels you run and what worries you today about producing them. We will reply with a proposal or with a reasoned “not yet”. [Pre-register](https://claqon.com/en/pre-registration/#configurador) [See all use cases](https://claqon.com/en/use-cases/)