Japanese agency group Dentsu has announced it has signed the European Union’s Code of Practice on Transparency of AI Generated Content, claiming to be the first global marketing group to do so. Dentsu says the move reinforces its commitment to responsible AI adoption, and helps establish a new standard for transparency across the ad industry.
“AI is creating extraordinary new opportunities for brands, but real value can only be realised when it is built on trust,” said Will Swayne, global practice president, media and integrated solutions at Dentsu. “By signing the EU Code of Practice on Transparency of AI Generated Content, we are helping clients navigate a rapidly evolving landscape with greater confidence, providing clear frameworks, governance and transparency around how AI generated content is created and used. This is about moving beyond compliance to enable responsible innovation at scale, helping clients unlock the benefits of AI while maintaining the trust of the people they serve.”
Larger transparency questions remain
The code is a voluntary framework for ensuring and demonstrating compliance with the EU AI Act, which came into effect earlier this month. It sets out rules for marking and detection of AI-generated and manipulated content, both for providers of AI systems (i.e., the big AI companies themselves), and for deployers — so businesses like Dentsu which use those AI systems.
The principles of the Act apply to all businesses, whether they sign up to the code of practice or not. For deployers, they specifically target deepfakes — that is, content which is AI generated or manipulated which resembles real people, objects, places, entities, or events, and which might falsely appear authentic or truthful. Deepfakes must be explicitly labelled as AI-generated or AI-modified. AI-generated text relating to matters of public interest meanwhile must run through human review before being published.
But the code in a sense represents the gold standard in terms of compliance, detailing when and where AI labels appear. And Dentsu says it will continue to work with regulators, technology partners, clients, and industry peers to create consistent approaches to transparency, and help develop industry-specific standards. The Japanese group looks keen to signal that it’s leaning in to AI transparency, rather than waiting for legal rulings to make clear the minimum requirements.
But the code covers just a few specific areas. There are larger questions for agencies around flagging the use of AI in generated assets which aren’t classed as deepfakes — and where they’re therefore not required to disclose under the EU AI Act. And then there’s the question of transparency for clients around exactly how and when AI is used throughout the marketing process, from ideation through to creative production and media buying.
Follow VideoWeek on LinkedIn.
