AI Regulation Kicks Off in Europe: Supervision and Penalties commence for AI Oversight
In the realm of artificial intelligence (AI), the European Union (EU) is taking a significant step forward with the active implementation of the AI Act. As of August 2, 2025, providers of AI systems will face official scrutiny from national authorities, marking a milestone in the regulation of AI technology within the EU.
The AI Act, which came into effect earlier in 2025, now includes provisions for penalties for breaches. From August 2, EU member states are required to notify the European Commission about their market surveillance authorities charged with supervising AI suppliers' compliance with the AI Act. This notification is the first formal step toward national oversight enforcement.
However, as of early August 2025, at least half of the member states have not yet appointed these national regulators. Delays are attributed to political factors such as recent elections, raising concerns about the readiness to enforce the rules effectively from August 2, 2025.
It is crucial that national authorities are appointed as soon as possible, and that they are competent and adequately resourced to oversee AI risks, including those affecting fundamental rights. The European Commission has received some notifications, but many are still pending consideration, and it is expected that several member states will miss this initial deadline.
Enforcement powers exist and will be coordinated at EU level. Market surveillance authorities can require corrective actions, order the withdrawal of non-compliant AI systems, and impose penalties. Enforcement will also be supported by an AI Board and scientific panel to ensure consistent application across the EU.
The overall EU timeline confirms no delay in the AI Act's milestones. Basic prohibitions and governance provisions took effect earlier in 2025, general-purpose AI model obligations came into force on August 2, 2025, and full high-risk AI system obligations and harmonized standards will apply by August 2, 2026.
The complexity of the AI Act becomes even more challenging when considering how multiple bodies at both EU and national levels need to coordinate together. There may be potential for overlaps and conflicts between the AI Act and existing regulations like the General Data Protection Regulation (GDPR), the Digital Services Act, and the Digital Markets Act.
As the AI Act enters its active implementation phase, the focus now shifts to the timely action of member states to set up their regulators and sanction regimes. The rules on general-purpose AI systems, including large language models, have entered into force. Companies may be fined up to €35 million or 7% of total worldwide annual turnover, whichever is higher, for breaches of the AI Act. For smaller companies, lower fines will apply.
In conclusion, the AI Act's implementation marks a significant step forward in the regulation of AI within the EU. However, the timely appointment of competent and adequately resourced national authorities remains critical to ensure effective enforcement. The EU's coordinated approach will provide a framework for consistent application of the AI Act across its member states.
- As the AI Act's implementation proceeds, the spotlight is on EU member states to expedite the appointment of national authorities, as their competence and resources are crucial for enforcing AI policy-and-legislation, particularly with regards to general-news about AI technology.
- The EU's General Data Protection Regulation (GDPR), Digital Services Act, and Digital Markets Act could potentially overlap or conflict with the AI Act, making it imperative for the multiple regulatory bodies at both EU and national levels to coordinate closely in the technology sector.
