AI Notice

This notice follows the transparency obligations of the EU Artificial Intelligence Act (2024/1689) and good practice.

1. Where we use AI

• In-site assistant and tools: question explanations, topic explanations, report drafts and language editing, presentation and academic support, feedback triage, news and literature summaries. Provider: Anthropic (Claude).
• Content improvement: generating improvement suggestions from usage data and feedback.
• Connecting your own Claude account (connector): here you use the AI service through your own account and its terms apply.

2. How to recognise it

AI-generated text is labelled "AI draft" or "AI summary". Screens where you talk to an AI say so clearly.

3. Limits and safeguards

• AI can make mistakes. Check important information against the source shown.
• Numerical calculations (doses, statistics) are not estimated by AI; they are computed with exact formulas.
• AI shows only verified references and does not invent sources.
• Medical content changes are never published automatically by AI; they are approved by an editor.
• Patient identifiers are never sent to AI; they are removed in the browser beforehand.
• Content sent to the AI provider is not used to train its models (to be confirmed by contract).

4. Academic use

If you prepare a thesis, article or presentation with AI support, disclose AI use according to the rules of the venue (e.g. ICMJE recommendations). The Platform reminds you.

5. Personalisation and objection

A personal style profile is created only with your consent. You can turn off AI suggestions and delete your profile. No decision with significant effects on you is made solely by automated processing.