|

The European Commission and the Artificial Intelligence Board have confirmed the adequacy of the Code of Conduct on the transparency of content generated by artificial intelligence

The European Commission and the Artificial Intelligence Board (AI Board) have adopted and published an adequacy assessment ofthe Code of Practice on Transparency of AI-generated Content. The final version of the Code was published on 10 June 2026.

Following this positive assessment, the Code has been officially recognised as an appropriate voluntary tool enabling providers and operators of certain artificial intelligence systems to demonstrate compliance with the obligations set out in Article 50(2), (4) and (5) of the Artificial Intelligence Act.

The Code and the European Commission’s Guidelines

The implementation of the rules on the transparency of content generated or adapted by artificial intelligence is supported by two key documents: the Code of Conduct and the European Commission’s Guidelines on the implementation of transparency obligations.

The Code of Conduct on the transparency of content generated by artificial intelligenceis a document to which providers and operators of artificial intelligence systems may voluntarily adhere. It focuses on the practical and technical implementation of the obligations set out in Article 50(2), (4) and (5) of the Artificial Intelligence Act.

It is intended in particular for providers and operators of artificial intelligence systems who:

• create or manipulate synthetic image, audio, video or text content;
• create or adapt deepfakes;
• publish AI-generated or AI-manipulated texts intended to inform the public on matters of public interest.

By adhering to the Code and implementing its provisions, signatories can demonstrate compliance with the aforementioned obligations under the Artificial Intelligence Act.

The European Commission’s guidelines on the transparency obligations under the Artificial Intelligence Act, which are currently still at the draft stage, primarily serve an interpretative purpose. The draft was open for public consultation until 3 June 2026.

The guidelines clarify the full legal scope of the obligations under Article 50 of the Artificial Intelligence Act, including obligations for systems not covered by the Code. These include, for example:

• artificial intelligence systems that interact directly with natural persons, such as virtual assistants and automated telephone systems;
• emotion recognition systems;
• biometric categorisation systems.

The guidelines also clarify who is a controller under Article 50, the scope of individual obligations, and how the rules and exceptions should be interpreted in practice. They address, for example, the distinction between deepfakes and the ordinary editing of synthetic content, as well as the issue of editorial responsibility.

The Code and the Guidelines therefore complement one another: the Code sets out practical ways of fulfilling certain obligations, whilst the Guidelines provide a broader explanation of their legal content.

What does a positive assessment of adequacy mean?

Following a positive assessment by the European Commission and the Artificial Intelligence Board, the Code has become an officially recognised tool for demonstrating compliance with transparency rules.

Providers and operators of artificial intelligence systems may rely on it to demonstrate compliance, regardless of their place of establishment, place of business or the competent market surveillance authority.

However, simply signing the Code does not automatically establish a presumption of compliance. Signatories will have to demonstrate that they are actually implementing its provisions. However, by fulfilling the commitments set out in the Code, they will be able to demonstrate compliance with the obligations under Articles 50(2) and 50(4) of the Artificial Intelligence Act in a more consistent and predictable manner.

Data controllers who do not sign the Code will be able to fulfil these obligations using their own solutions. However, they will have to demonstrate to the supervisory authorities that their solutions meet the required standards of effectiveness, robustness, interoperability and reliability. Such demonstration may be more demanding and may entail more detailed scrutiny by the competent authorities.

What does the Code set out?

The Code sets out practical and technical methods for labelling and identifying content created or modified by artificial intelligence.

The measures envisaged include, for example:

• protected and digitally signed metadata;
• watermarks;
• standardised icons;
• techniques for detecting content created or modified by artificial intelligence.

The Code is divided into two sections.

The first section is intended for providers of artificial intelligence systems. It contains commitments regarding machine-readable labelling and enabling the detection of synthetic content that has been created or modified by artificial intelligence.

The second section is intended for users of artificial intelligence systems. It contains rules for labelling deepfakes and texts created or modified by artificial intelligence that are published with the aim of informing the public about matters of public interest.

For the purposes of the second section of the Code , a set of ‘AI’ icons has also been developed , which implementers can use to label content created or modified by artificial intelligence.

Organisations may adhere to both sections of the Code or only to the one relevant to their role and activities.

Why is the Code important for data controllers?

For providers and implementers of artificial intelligence systems, the Code primarily offers a clearer and more standardised way of demonstrating compliance.

Signatories who actually implement its requirements will be able to demonstrate compliance with obligations regarding transparency and content labelling by referring to their adherence to the provisions of the Code. The competent authorities will still verify compliance, but their checks may focus primarily on whether the signatory is honouring the commitments made.

Adherence to the Code and compliance with it may also be taken into account as a mitigating factor when determining the amount of any fine.

Nevertheless, the Code is not the only possible means of ensuring compliance. Obligated parties may opt for their own technical and organisational solutions, but they will have to demonstrate that these achieve a comparable level of effectiveness and reliability.

The European Commission has also published a list of frequently asked questions for potential signatories and other interested organisations, which provides further clarification on the content of the Code, the accession process and the implementation of individual commitments.