The EU AI Act has set a global benchmark for regulation, but real-world deployment has exposed gaps and implementation issues that regulators are still working through.
After one year of enforcement, the law has driven stronger governance in high-risk systems while also revealing practical friction points for smaller companies and cross-border operations.
1. What worked
Mandatory risk assessments, transparency requirements, and post-market monitoring have helped bring structure to AI governance and given users clearer expectations about high-risk systems.
2. What failed
Compliance costs became barriers to entry for smaller companies. The vagueness around "high-risk" applications created confusion. And enforcement has been slower than expected, with regulators still building capacity.
3. The unintended consequences
Some companies have moved development outside the EU to avoid compliance overhead. Others have simply added documentation without changing underlying practices. Enforcement gaps have allowed both strategies to continue.
"The EU AI Act is better than no regulation, but it's also clear that regulation written before the technology matured creates as many problems as it solves."
4. What's coming next
The Commission is working on clarifications for the ambiguous definitions. Harmonization guidelines are being developed. And international bodies are studying how to align with or diverge from the EU framework.
The next two years will determine whether the AI Act becomes a global standard or a uniquely European constraint on innovation.