EU AI Act: New Rules for a New AI Landscape
As of today, August 2, 2025, the European Union’s landmark AI Act has begun to roll out, bringing with it the first legal framework of its kind in the world. This comprehensive legislation is designed to create a “human-centric and trustworthy” environment for artificial intelligence by addressing risks, ensuring transparency, and protecting fundamental rights. The new rules, particularly for general-purpose AI models like ChatGPT and Gemini, mark a significant step in the global governance of AI technology.
A Risk-Based Approach to Regulation
The EU AI Act is founded on a tiered, risk-based approach that classifies AI systems into four categories:
* Unacceptable Risk: AI systems that pose a clear threat to people’s safety, rights, and livelihoods are outright banned. This includes practices like social scoring by governments, using subliminal techniques to manipulate behavior, or exploiting vulnerabilities based on age or disability.
* High-Risk: These systems, while not banned, are subject to stringent requirements. They are used in sensitive areas such as medical devices, critical infrastructure, law enforcement, and employment. Providers of high-risk AI must undergo a conformity assessment and meet strict obligations related to data quality, technical documentation, human oversight, and robustness.
* Limited Risk: Systems like chatbots and deepfake generators fall into this category. They are required to meet specific transparency obligations, such as informing users that they are interacting with an AI system or that the content they are seeing has been artificially generated.
* Minimal or No Risk: The vast majority of AI systems, such as spam filters or video games, are considered to be of minimal or no risk and are largely unregulated by the Act.
New Transparency Mandates for General-Purpose AI
The rules that came into force today specifically target general-purpose AI (GPAI) models. These are systems with a wide range of applications, and the new requirements are designed to bring a new level of accountability to their development. Providers of these models are now obligated to:
* Disclose Training Data: They must publicly report a summary of the copyrighted material used for training their models, including which sources they scraped from the internet. This aims to give creators greater insight into how their work is being used.
* Establish Contact Points: Developers are now required to set up a dedicated point of contact for copyright holders, making it easier for artists, writers, and publishers to communicate with the companies and address potential infringements.
* Document Safety Measures: For the most powerful GPAI models that could pose a systemic risk to the public, there is an additional requirement to document and report on the safety measures and risk mitigation strategies they have in place.
The Copyright Debate and Future Enforcement
While these measures are a significant win for copyright holders, some associations of creators and publishers have voiced criticism, arguing that the rules don’t go far enough. They contend that without a mandate to name specific datasets or sources, the transparency obligations remain “ineffective” and do not provide sufficient protection for intellectual property.
Enforcement of the EU AI Act will be carried out by the newly established European AI Office and will be phased in over the coming years:
* Unacceptable risk AI systems have been banned since February 2025.
* General-purpose AI rules apply as of today, August 2, 2025.
* High-risk AI systems will have to comply with the rules starting in August 2026.
* The European AI Office will begin enforcing rules for legacy AI models (those placed on the market before August 2, 2025) starting in August 2027.
Violations of the Act can result in severe penalties, with fines reaching up to 35 million euros or 7% of a company’s total global annual turnover, whichever is higher, for banned AI systems. Other violations can lead to fines of up to 15 million euros or 3% of turnover.
The EU AI Act is a global first that has already influenced AI policy discussions around the world. It sets a precedent for how governments can regulate a rapidly evolving technology to ensure it is developed and used responsibly, with human safety and rights at its core.
Full Text of the EU AI Act
The full legal text of the EU AI Act, officially known as Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), can be accessed through the official database of EU law:
Regulation (EU) 2024/1689 on EUR-Lex
https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng
* His Highness and Excellency Global Peace Ambassador Dr.iur. PRINCE MARIO-MAX SCHAUMBURG-LIPPE is a Lawyer, working Royal, award winning TV- and Event-Host, Bertelsmann Randomhouse author, Public Speaker and Philanthropist ♔ Prince Mario-Max is the son of Royal Dignitaries H.H. Dr.h.c. Prince Waldemar and H.H. Dr. Princess Antonia of Schaumburg-Lippe. His Grandmother is H.R.H. Princess Feodora of Denmark. Therefore they are the Royal Danish Nachod Line of The Princes of Schaumburg-Lippe, the Founding Family of Hamburg, Lübeck and Kiel. ♔ Instagram https://www.instagram.com/princemariomax/ Website http://www.schaumburglippe.org Facebook https://www.facebook.com/zuschaumburglippe X-Twitter https://twitter.com/schaumburglippe Linkedin https://www.linkedin.com/in/prince-mario-max-schaumburg-lippe-1879978a
