On 9th July, the Commission published its opinion finding that the Code of Practice on Transparency of AI-generated content adequately covers the AI Act transparency obligations under Articles 50(2), (4) and (5).

As a reminder, the voluntary code is intended to help providers and deployers of generative AI systems comply with rules on marking, detecting and labelling AI-generated or manipulated content, in particular deepfakes and certain public-interest text.

The AI Board also assessed the code as adequate, noting that it sets out eight commitments covering obligations for both providers of AI systems and deployers using AI systems to generate deepfakes or certain text.

Providers and deployers may rely on the code to demonstrate compliance across the EU, but adherence does not constitute conclusive evidence of compliance, and competent authorities may still assess actual implementation.

We have been informed that the transparency guidelines are expected to be published around 20 July.