Europe's long-stalled AI Act implementation is moving forward with the constitution of two critical oversight bodies and the publication of transparency rules for AI-generated content. The European Commission announced the members of the Advisory Forum and Scientific Panel in June 2026, marking a major milestone after significant delays. These bodies will shape how the EU's landmark AI regulation actually works in practice.

What Are These New Oversight Bodies, and What Do They Do?

The two oversight bodies have distinct roles in the AI Act ecosystem. The Scientific Panel consists of 60 independent experts tasked with actively overseeing General-Purpose AI (GPAI) models, which are large AI systems like ChatGPT or Claude that can be adapted for many different tasks. The Advisory Forum, with 172 members, functions as a broader technical advisory body with balanced representation from commercial and non-commercial sectors, providing targeted advice to the European Commission and the AI Board.

The distinction matters because it reflects Europe's attempt to balance innovation with oversight. The Scientific Panel acts as a watchdog; the Advisory Forum acts as a sounding board. Together, they represent the first real institutional machinery for enforcing the AI Act at scale.

What New Transparency Rules Are Now in Effect?

Alongside the oversight bodies, the European Commission published the final Code of Practice on marking and labeling of AI-generated content, operationalizing the AI Act's transparency requirements. The Code is split into two main sections:

• Provider Rules: Generative AI system providers must mark and enable detection of AI-generated and manipulated content.
• Deployer Rules: Organizations using AI systems must label deepfakes and AI-generated or manipulated published text.

The Code itself is tiered: some measures are mandatory for compliance, others are voluntary but recommended, and some are purely voluntary. OpenAI became the first major actor to publicly announce it will sign the Code, signaling that at least some leading AI companies are willing to adopt these transparency standards.

How Is Europe Addressing Its Competitive Disadvantage in AI?

The oversight bodies and transparency rules are only part of Europe's response. In June 2026, the European Commission published its Tech Sovereignty Package, a comprehensive strategy to reduce Europe's dependence on foreign AI providers and infrastructure. The package identifies the core challenges Europe faces:

• Technological Dependencies: Europe relies heavily on foreign providers for critical AI infrastructure and cloud services.
• Market Concentration: A small number of companies control most of the AI market, creating vendor lock-in risks.
• Infrastructure Gaps: Europe lacks sufficient data center capacity and semiconductor production to support large-scale AI development.

The Commission's response includes two major legislative proposals: the Cloud and AI Development Act (CADA), which aims to triple the EU's data center capacity within five to seven years by speeding up environmental assessments, and Chips Act 2.0, which seeks to boost and diversify Europe's semiconductor supply chain.

However, some observers argue that regulation alone is insufficient. Dr. Sergey Lagodinsky, Vice Chair of the Greens/EFA Group in the European Parliament, warned that Europe risks becoming "a digital colony between two AI empires" without more aggressive action. He emphasized that the U.S. demonstrated an "AI kill switch" when it restricted access to Anthropic's Mythos frontier model, potentially cutting off 88 percent of the world's population from advanced American AI systems overnight.

Sergey Lagodinsky, Vice Chair of the Greens/EFA Group in the European Parliament

What Steps Must Europe Take to Compete?

Experts and policymakers have outlined several priorities for Europe to strengthen its AI position:

• Scale Investment: Europe must mobilize private capital at the scale of trillions, not billions, to fund AI infrastructure and research. Public funding alone cannot close the gap with the U.S. and China.
• Unified Market Strategy: National solutions will not work. Europe must leverage its common market through initiatives like French-German AI gigafactories and European space constellations for compute capacity.
• International Partnerships: Europe should pool AI compute capabilities with middle-power nations through multilateral consortiums, positioning itself as the only superpower willing to use multilateralism to counter digital imperialism.
• Strategic Cooperation with the U.S.: In the medium term, Europe will depend on American compute and chip infrastructure. Europe should build a relationship based on regulatory standards and trust, while leveraging assets like Dutch chip-printing technology from ASML.

Lagodinsky argued that regulatory standards can actually strengthen Europe's position by anchoring American technologies in European trust and transparency, creating a foundation for strategic parallelism: "cooperating with Washington where we must, while building capacities and infrastructure that will one day let us stand alone".

Lagodinsky

What's Next for EU AI Governance?

Starting July 1, 2026, Ireland takes over the Presidency of the Council of the European Union, with a publicly released program focused on strengthening EU capacities in cloud computing and artificial intelligence. Ireland will host an AI Summit in Dublin on October 14, 2026, bringing together EU and global leaders, CEOs, investors, and academics to discuss harnessing AI for European competitiveness.

However, a major challenge looms: the Digital Omnibus proposal, a draft law seeking to amend various digital laws including the General Data Protection Regulation (GDPR). Member states failed to agree on a common position during the Cypriot presidency, and concerns have been raised that the Omnibus could relax data processing rules in AI development and operation. The Irish presidency's historical relationship with the tech industry and recent controversies over data protection enforcement have raised questions about how aggressively Ireland will push back against industry pressure to weaken privacy protections.

The tension is clear: Europe is trying to build AI competitiveness through both regulation and infrastructure investment, but the regulatory framework itself remains contested. The oversight bodies and transparency rules represent real progress, yet they exist in a landscape where Europe's compute power and capital remain far behind the U.S. and China. Whether Europe can thread this needle, competing globally while maintaining its regulatory standards, will define the next phase of AI governance.