AI Transparency Obligations Now Apply

The transparency obligations under article 50 of the EU's AI Act are now live. Basically, the aim is to identify AI-generated or manipulated content. Subject to certain exceptions, users must be informed when they are interacting with an interactive AI system, such as a chatbot; deepfakes must be clearly labelled, as must AI-generated or manipulated text that is published on matters of public interest. There is 4 month transitional period for machine-readable marking by AI systems first made available in the EU market before 2 August 2026. This post is for information purposes only. If you need advice, please let me know.

Meaning of 'Provider' and 'Deployer' etc

A ‘provider’ is:

a natural or legal person, public authority, agency or other body that: 

develops an AI system or a general-purpose AI model or 

has an AI system or a general-purpose AI model developed 

and 

places it on the market or  

puts the AI system into service under its own name or trademark, 

whether for payment or free of charge; 

A 'deployer' is: 

"a natural or legal person, public authority, agency or other body using an AI system under its authority except where the AI system is used in the course of a personal non-professional activity;"

‘AI system’ means: 

"a machine-based system that is 

designed to operate with varying levels of autonomy and

that may exhibit adaptiveness after deployment, and 

that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments;"

'general-purpose AI model’ means: 

an AI model, including where such an AI model is trained with a large amount of data using self-supervision at scale, that 

displays significant generality and 

is capable of competently performing a wide range of distinct tasks regardless of the way the model is placed on the market and 

that can be integrated into a variety of downstream systems or applications,

except AI models that are used for research, development or prototyping activities before they are placed on the market; 

What Must Providers and Deployers Do? 

There are certain useful carve-outs or exceptions, but generally:

Providers must:

design AI systems that interact directly with individuals (including chatbots) to disclose that users are engaging with an AI system. 

mark AI-generated or manipulated audio, image, video or text in a machine-readable format, using effective, interoperable, robust and reliable technical solutions for detection. 

Deployers must:

inform individuals when they are subject to emotion recognition or biometric categorisation. 

disclose when some content (deep fakes or text on matters of public interest) has been artificially generated or manipulated.

'deep fake’ means AI-generated or manipulated image, audio or video content that resembles existing persons, objects, places, entities or events and would falsely appear to a person to be authentic or truthful;

EU Voluntary Code

Provides and deployers can sign up to an EU voluntary code that has more guidance on achieving and assessing AI transparency here. 

This post is for information purposes only. If you need advice, please let me know.