Synthetic content marking policy

Purpose: to describe how ROIFORCIO GmbH marks AI-generated and AI-manipulated outputs so that they are detectable as artificial, in accordance with Article 50(2) of the AI Act. This is both an external statement and an internal engineering standard.

1. Scope

This Policy applies to all outputs generated by the Service that constitute synthetic audio, image, video or text, including: AI Presenter videos and voice-overs; AI avatars; AI voice cloning outputs; and AI video translation and dubbing.

2. Marking standard

  • Outputs are embedded with machine-readable provenance markers using a recognised standard (e.g. C2PA content credentials) and, for audio, a robust audio watermark.
  • Markers identify the content as artificially generated or manipulated and survive common transformations (re-encoding, compression) as far as technically feasible.
  • The solution is designed to be effective, interoperable, robust and reliable, taking into account the state of the art, the type of content and the cost of implementation.

3. Third-party generative components

Where generation is performed by a third-party GPAI provider, ROIFORCIO GmbH ensures that the final output delivered through the Service carries the required marking, either by relying on provider-side marking or by applying its own marking layer at the point of export.

4. Visible disclosure for deepfakes

For content resembling real persons, the Service additionally supports a visible label (e.g. an “AI” indicator) that deployers can and, where Article 50(4) applies, must present to the audience.

5. Exceptions

Machine-readable marking does not apply to the extent the AI system performs an assistive function for standard editing or does not substantially alter the input data provided by the deployer or its semantics.

6. Governance and timing

Item

Position

Owner

Engineering, with Legal oversight

Applicable date – new systems

2 August 2026

Applicable date – systems on market before 2 Aug 2026

2 December 2026 (Omnibus grace period)

Reference standard

Code of Practice on Transparency (final version expected mid-2026)

ROIFORCIO GmbH will assess adherence to the EU Code of Practice on the marking and labelling of AI-generated content, which provides a presumption of conformity with Article 50(2).