The European Union has entered the implementation phase of the Artificial Intelligence Act, a package of rules aimed at setting limits on the riskiest uses of the technology and requiring greater transparency for AI-generated content. At the center of attention are deepfakes, high-risk systems and the fines that could hit companies that break the rules.
What changes with the rules coming into force
The main rules of the European Artificial Intelligence Act began to be enforced from the start of August across the EU. The approach of this legal framework is based on risk: some practices are banned, while systems used in sensitive areas face stricter obligations.
Under the new framework, some content created or manipulated by artificial intelligence must be identifiable. In some cases, users must be informed when they are interacting with AI, a requirement that directly affects the debate over digital manipulation and online fraud.
Deepfakes and fines for companies
One of the most important points concerns deepfakes, at a time when distinguishing between real material and fabricated material is becoming increasingly difficult for the public. The new rules require such content to be identifiable, in an effort to limit abuse of tools that can affect public opinion, privacy and security.
For companies that fail to comply with the obligations, strong financial sanctions are предусмотрены. Depending on the violation, fines can reach up to 15 million euros or 3% of global annual turnover, showing that Brussels is trying to turn regulation from a political declaration into a real mechanism of pressure.
Majlinda Bregu’s reaction
Former Integration Minister Majlinda Bregu also reacted to this development, viewing the European framework as an effort to prevent harm rather than intervene after the consequences have become irreversible.
In her reaction, Bregu stressed that the protection of children and adolescents in the digital environment should not be conditioned by a country’s political status toward the EU. According to her, the same standard of protection that applies to a teenage girl in Paris or Berlin should also apply to a 14-year-old girl in Tirana.
Why the debate also affects Albania
Although the law is enforced in the EU, its impact goes beyond the borders of the bloc, especially for countries such as Albania that use the same platforms, applications and digital environments. This makes the debate less theoretical and more a concrete issue of security, media literacy and the protection of minors.
The main message emerging from this development is that the use of AI cannot be left solely in the hands of the market or the pace of the industry. If technology moves faster than the law, the question remains how ready Albanian institutions are to build minimum standards of protection, instead of simply waiting for moves from Brussels.
The European Artificial Intelligence Act is being presented as the first serious attempt to set limits on a technology that is rapidly entering everyday life, education, information and public communication.
Beyond the rhetoric about innovation, the real challenge remains whether the rules will be enforced effectively and whether the protection of citizens, especially children, will be treated as a priority rather than as an issue endlessly postponed.