Thus squeezes the nuts the EU’s law to the agencies and the marketing departments

The ending becoming each day’s bread in the universe of marketing and advertising. However, The framing “boom” of fashion technology in this area of activity will now be ahead subject to strict regulations in the legal level with the European Law on Artificial Intelligence (AI ACT). Some parts of the Law of the European Union (EU) entered into force during this weekend, which inevitably advocates advertising agencies and marketing departments to reexamine the way they use AI (if they don’t want to be brew perhaps with possible fines).

If the General Data Protection Regulation (RGPD) of the EU has already regulated the processing of user personal data, The new law of AI goes a step further, put on the spotlights not only what data is processed but also the way in which such data is processed.

By virtue of the European Artificial Intelligence Law, The AI systems are divided into four different classes depending on the risks involved. Most of the AI applications that are used in the marketing and advertising field are attached to the category of limited nature risks (the use of chatbots or automated text generation, for example). However, this does not exempt such applications of scrupulously certain requirements directly related to transparency.

Since August 2, 2025, who believe with the support of the content that can be misrepresented as real (“Deepfakes”, for example) must clearly label it as such. And also companies that contact the client based on automation tools must make clear clearly pristine that in their communications with the consumer the AI is involved. The days when brands could experience quietly with AI safe from any legal restriction have arrived, therefore, to an end.

In accordance with article 5 of the EU’s law, The techniques that involve the use of AI with the last objective of manipulating people are also prohibited sharply. In this sense, brands cannot be supported at any time in applications that deliberately exploit the weaknesses of their customers or in algorithms that manipulate the “reviews” that are reached in the eyes of their clientele.

The law of the EU forces companies to reexamine all the processes in which the AI is directly involved

Companies cannot qualify their customers based on data on their behavior and credit ratingmore reason for companies to look with magnifying glass from now on the practices in which the AI is directly involved.

On the other hand, to note, what The content that is generated only and exclusively with the support of AI will be considered in the public domain, which inevitably raises a dilemma for the marketing agencies and departments that work with that type of content. In practice, this precept of EU’s law means that those who generate content without more support than artificial intelligence will not in any case hold exclusive rights over that content.

You must not forget that The new EU directive on responsibility for damages caused by defective products (which will be implemented from next year) It ultimately responsible for companies if their AI systems are of a defective nature. According to that directive, the AI systems are “products” and, as such, the errors that may lead to the direct responsibility of those who are lightened and marketed. To ensure that their AI systems are not at the mercy of errors (something that inevitably involves much complexity), companies must rely on the “human in the loop” methodology (Hitl), which directly integrates human interactions in the process of development of artificial intelligence models.

To avoid shocks (which can be translated eventually in fines) with the entry into force of the EU’s law, Experts recommend companies to leverage four aspects mainly:

1. Strategy and selection of tools. It is necessary to establish the objectives with which certain AI tools are used and also corroborate if the chosen supplier meets those provided in the GDPR.
2. Legal review. Companies must have clearly clear what is the risks of the AI tools that they use in their day to day, and determine if such tools process personal data and make it scrupulously respecting the law.
3. Design process. It is essential that companies entrust humans of flesh and blood to supervise the results on the table by AI and that they adequately train employees who use this technology.
4. Operation and monitoring. Companies must ensure that the processes in which the AI is properly documented and that quality exams are continuously undertaken.

The entry into force of the EU’s law forces the agencies and marketing departments to use artificial intelligence much more mature and abandon to some extent the experimentation stage (in which to act negligently in this area of activity had no legal consequences). It is not just about knowing the tools of AI, but also of understanding the legal implications of such tools, whose success in the area of marketing and advertising can no longer be measured only and exclusively in the form of clicks, but also attending to its ethical and legal employment.