Critics of the current implementation strategy warn that the repeated delays and the introduction of the Digital Omnibus have transformed what was once a landmark piece of legislation into a potential 'paper tiger.' By pushing back the deadlines for high-risk AI systems—which include technologies used in sensitive areas like hiring, credit scoring, and law enforcement—the EU risks leaving a significant regulatory vacuum that could persist for years.
Accountability advocates argue that these concessions to industry pressure undermine the original intent of the AI Act: to protect citizens from the most dangerous applications of the technology. While transparency labels for chatbots are a welcome development, they do little to mitigate the systemic biases or fundamental rights violations that can occur within high-risk AI models. The delay effectively grants a grace period to powerful tech companies, allowing them to continue deploying potentially harmful systems without the rigorous oversight originally promised.
Moreover, the complexity of the current, fragmented timeline creates confusion for both businesses and the public. As the rules for different systems take effect at different times, the lack of a unified, immediate enforcement mechanism may lead to inconsistent application across EU member states. For those concerned with the rapid, unchecked expansion of AI, the current state of the AI Act suggests that political expediency is being prioritized over the urgent need for comprehensive, enforceable safeguards.