How the EU AI Act Is Reshaping Tech Companies in 2026
The European Union’s Artificial Intelligence Act, which came into full force in August 2024, is now entering its most consequential phase. As of mid-2026, the compliance deadlines for high-risk AI systems have kicked in, and companies operating in the European market are scrambling to adapt — or facing significant penalties.
Under the Act, companies deploying AI systems classified as “high-risk” — including those used in hiring, credit scoring, medical diagnosis, and critical infrastructure — must meet strict transparency, documentation, and human oversight requirements. Fines for non-compliance can reach up to €35 million or 7% of global annual turnover, whichever is higher.
Compliance Reality Sets In
The first six months of enforcement have revealed a mixed picture. Large technology firms with dedicated legal and compliance teams — think Google, Microsoft, and SAP — have largely achieved compliance, though not without significant investment. Microsoft alone reportedly spent over €400 million on EU AI Act compliance measures in 2025.
For small and medium-sized enterprises (SMEs), the picture is more challenging. A survey by DigitalEurope found that 43% of European SMEs developing AI products are not fully compliant, citing the complexity of the requirements and the cost of external legal counsel.
Innovation vs. Regulation
The debate over whether the AI Act stifles innovation or provides necessary guardrails has not subsided. Proponents point to increased consumer trust metrics in AI-powered services across the EU. Critics argue that the regulatory burden is pushing startups to launch first in the US or Asia before entering the European market.
“The AI Act is the most comprehensive AI regulation in the world, and it’s setting a global precedent,” said a policy analyst at a Brussels-based think tank. “But the implementation needs to be more startup-friendly. The regulatory sandboxes help, but they’re not a complete solution.”
What Companies Should Do Now
Legal experts recommend that companies take three immediate steps: conduct a thorough inventory of all AI systems in use or development, classify each system according to the Act’s risk categories, and implement documentation workflows that can demonstrate compliance to regulators on demand. The European Commission has also published simplified compliance guidelines specifically for SMEs, available through the EU’s AI Office portal.
As the enforcement landscape matures, one thing is clear: the EU AI Act is not just a European concern. Any company anywhere in the world that wants to serve European customers must play by these rules — making it a de facto global standard for responsible AI development.
The Netherlands, for internationals
One email a week: the week's news for internationals in the Netherlands, new practical guides and what changed in the cities. No spam, unsubscribe any time.
