Illustration and visual intention

How can you obtain an illustration suited to professional communication materials and check its quality before publication?

1250 € per session, excl. VAT

In short

In a half-day session of 3 hours discover how to guide generative AI to produce an illustration that meets a simple brief. You will practise formulating instructions, comparing variants and checking the final file, with attention to rights, consent and publication notices.

Why this training?

An attractive image can still be unusable if it contradicts the message, overcrowds the poster or leaves too little room for text. Describing the subject, composition and medium turns a visual idea into verifiable instructions. Generation can produce inconsistent details, unwanted resemblances or elements whose rights need to be examined. Comparing versions and documenting limitations helps you select an illustration without confusing aesthetic quality with suitability for publication. Continuing case study: The fictitious association « Avelune Partage » is preparing a poster for a bicycle repair workshop. It wants a warm illustration showing two adults around a bicycle, with no visible branding and a clear area for adding the title and practical information later. Transfer to the workplace: For your next communication material, reuse the brief and checklist. Test the illustration at its actual intended size, then have the rights and notices approved before any publication.

Objectives of this training

  • Set out a brief specifying the message, audience, medium and visual constraints.
  • Write a structured, clear generation prompt.
  • Compare two variants in terms of composition and readability.
  • Identify visual inconsistencies and risks relating to representation.
  • Export an illustration and complete its checking record.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communications officers who are new to generating illustrations.
  • Assistants and association managers preparing posters or presentation visuals.

Prior skills

  • No prior AI knowledge required.
  • Be able to save an image file and find a document on your computer.

What you will work on

  1. Understand the need and define the image

    Objective: Translate the poster request into a visual brief. Illustration versus photography; main message, framing, palette, contrast and space for text. Limitations of a generated image.

  2. Formulate and test a prompt

    Objective: Obtain comparable initial proposals. Prompt structure: subject, action, environment, composition and rendering style. Variability of results; embedded text is often imperfect and should be added separately.

  3. Observe and improve the proposal

    Objective: Choose a version based on visible evidence. Visual hierarchy, object consistency, anatomy, stereotypes and distracting details. Checks at a small size and when enlarged.

  4. Prepare for responsible publication

    Objective: Export the visual and identify the approvals still required. Dimensions, format and readability after export. Rights to reference materials and permitted uses; consent from recognisable people; notices appropriate to the context. Deepfakes and misleading representations must not be used.

What you take away

  • A one-page visual brief for the fictitious poster.
  • The initial prompt and an adjusted version, together with the resulting variants.
  • The selected illustration, exported for layout.
  • A completed quality checklist and publication note.

Method and assessment

  • Short teaching inputs followed by immediate practice on the continuing 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: Arrange access to a tool authorised by your organisation, with image generation available and accessible terms of use. Prepare a fictitious or anonymised brief; bring only visual reference materials that you are authorised to use.

Observable criteria: The brief states the message, audience, medium and area reserved for text. ; The prompt specifies the subject, composition, palette and elements to avoid. ; Two variants are compared against identical criteria, with a justified selection. ; The image is inspected at its intended size: bicycle, hands, silhouettes and absence of branding. ; The note distinguishes between verified rights, consent where applicable, required notices and outstanding reservations..

Frequently asked questions

Do you need to be able to draw?

No. The exercises focus on describing an intention and observing the result, without technical drawing.

Can you use a photograph of a colleague?

Not during this exercise. For subsequent use, check consent, rights and internal rules before uploading anything.

Is the resulting image automatically royalty-free?

No. You must examine the service’s terms, the references used and any recognisable elements before publication.

Is the file ready to send straight to the printer?

Not necessarily. The exported file must be checked against the printer’s specifications; additional graphic preparation may still be needed.

Last updated: