EU AI Act Under Scrutiny: Can Regulation Keep Pace with Emerging AI Risks?

inclusiveintellige

Oct 23

3 min read

As advanced AI systems evolve, European regulators and technology leaders are debating whether existing rules can respond to new risks while protecting fundamental rights and supporting innovation.

The European Union's Artificial Intelligence Act, adopted in 2024, established the world's first comprehensive legal framework for AI. Designed around different levels of risk, it aims to promote innovation while protecting safety and fundamental rights.

Yet as increasingly capable AI systems emerge, questions are growing about whether the existing framework can adequately address new technological developments.

A New Debate Over the AI Act

At an international AI summit in Dublin in October 2026, Microsoft President Brad Smith questioned whether existing regulations remain fully suited to rapidly evolving AI technologies.

Google DeepMind co-founder Demis Hassabis also raised concerns about the difficulty of legislating for technologies whose capabilities can change significantly within a few years.

European Commission officials defended the existing framework. Henna Virkkunen, the EU's technology commissioner, argued that the AI Act is equipped to address emerging technologies, including AI agents. Justice Commissioner Michael McGrath similarly expressed confidence in the current system while acknowledging the need for rules to remain adaptable.

The disagreement reflects a fundamental question in technology governance: should regulation be revised whenever new technological risks appear, or can existing principles be applied effectively to new systems?

From Legal Framework to Effective Oversight

The AI Act follows a risk-based approach, with stricter requirements for applications that may significantly affect people's safety or fundamental rights.

High-risk uses include certain AI systems deployed in employment, education, essential services, law enforcement and migration management. The relevant requirements cover risk assessment, documentation, transparency, human oversight and mechanisms to reduce discriminatory outcomes. Implementation, however, is taking place in stages.

According to the European Commission, key requirements for AI systems used in certain high-risk areas will apply from 2 December 2027, while requirements for high-risk systems integrated into regulated products will apply from 2 August 2028.

This means that some of the framework's most consequential protections have yet to be tested through full implementation.

Newer technologies, particularly AI agents capable of performing interconnected tasks and taking actions across different systems, also raise questions about accountability when multiple developers, providers and users are involved. While the AI Act establishes a regulatory foundation, its practical effectiveness will depend on technical standards, enforcement capacity and the ability of authorities to assess emerging risks.

Europe's Broader Challenge

The debate extends beyond regulation alone. CNN has highlighted Europe's difficulty competing with the United States and China in advanced AI development, including differences in investment, infrastructure and the availability of leading AI models.

These challenges raise a wider strategic question: how can Europe strengthen its technological capabilities while maintaining its commitment to fundamental rights?

Regulatory certainty and innovation are not necessarily opposing objectives. Clear rules can support trust and responsible technological development, but fragmented implementation and uncertainty about compliance may also create difficulties for businesses and public authorities.

Our Perspective: Governance Must Evolve with Technology

For Inclusive Intelligence International, the current debate illustrates the importance of regulatory frameworks that remain both adaptable and firmly grounded in human rights.

As AI systems become more autonomous and interconnected, questions of accountability, transparency and human oversight become increasingly significant. The AI Act provides an important foundation, but its effectiveness cannot be assessed solely by the existence of legislation. It must also be measured by whether institutions can enforce its requirements and whether people affected by AI systems have meaningful protections.