First video test

How do you turn a simple visual intention into a generated clip, then decide whether the result is usable?

1250 € per session, excl. VAT

In short

This half-day session of 3 hours offers a guided introduction to video generation. Working from a fictional brief, participants describe a scene, start a generation, revise their request and assess the output. They prepare a clip accompanied by its prompt and a pre-release checklist.

Why this training?

An appealing description is not enough to produce a coherent scene. Movements, objects and lighting can vary unexpectedly; learning to observe these discrepancies prevents you from confusing visual impression with usable quality. A first clip allows you to experiment without undertaking a complex production. Working on a single scene makes the effects of the prompt visible, while introducing rights checks and transparency about the synthetic origin of the images. Ongoing case study: The fictional organisation « Avelune Céramique » is preparing an internal presentation of its new cup. It wants a short shot showing a cup without a logo on a table, lit from the side, with a subtle camera movement. No faces, voices or music are needed. Transfer to the workplace: After the session, reuse the prompt for a different fictional object, change only the framing and compare the outputs using the same assessment grid. Before any professional use, have the rights, notices and distribution context approved by the appropriate person.

Objectives of this training

  • Describe a scene with a subject, setting, framing, lighting and movement.
  • Write a clear video prompt limited to a single action.
  • Start a generation in a tool authorised by your organisation.
  • Compare two outputs and identify their visible flaws.
  • Document rights, notices and reservations before release.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communications officers wishing to explore generated moving images.
  • Beginner content creators looking to test a simple visual intention.

Prior skills

  • No prior AI knowledge required; guidance is provided on using the tool.
  • Ability to use a browser and save a video file.

What you will work on

  1. Observe and define a scene

    Objective: Distinguish a feasible intention from an overly complex request. Principles of video generation, variability of results, duration and format according to the available options; the difference between subject movement and camera movement.

  2. Write and launch the first test

    Objective: Translate the brief into visual instructions. Prompt structure: subject, environment, framing, lighting, movement, constraints. Settings that are actually accessible and generation times.

  3. Compare and correct the output

    Objective: Improve a request based on precise observations. Cup distortions, unstable setting, flickering, excessive movement; criterion-by-criterion comparison rather than an overall aesthetic judgement.

  4. Prepare for responsible use

    Objective: Assess the clip’s suitability before sharing it. Rights relating to sources and terms of use for the result; consent for faces or voices; risks of deceptive deepfakes; notices of synthetic origin according to the context and applicable obligations.

What you take away

  • Visual brief for a single scene.
  • Initial prompt and revised version with the change explained.
  • Selected clip, exported if the tool’s capabilities allow.
  • Comparison grid and pre-release checklist.

Method and assessment

  • Short instructional inputs followed by immediate practice on the ongoing case study.
  • Individual work followed by work in pairs, with peer review of the outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Check access to a tool authorised by your organisation, its export options and the generation capacity available for the session. Prepare a fictional brief and, only if necessary, an anonymised visual with documented usage rights; no confidential data.

Observable criteria: The prompt specifies the subject, setting, framing, lighting and one movement. ; The correction addresses an identified variable and responds to an observed flaw. ; The grid distinguishes compliance with the brief, stability of shapes, continuity of movement and artefacts. ; The checklist records the rights for the elements used, any consent and the notices to be included. ; The selected output is watched in full and its limitations are recorded..

Frequently asked questions

Do you need to know how to edit a video?

No. The exercise focuses on a single shot, without complex editing or a soundtrack.

Can you use a colleague’s face?

Not in this exercise. Such use would require, among other things, appropriate consent, the necessary permissions and an assessment of the risk of deception.

Can the clip be published immediately?

Not automatically: rights, terms of use, notices and flaws must be checked before release is approved.

What should you do if generation is unavailable?

The analysis continues using prepared teaching outputs; the prompt remains ready to test, without promising an export during the session.

Last updated: