The EU's new AI law comes into force: What does it mean for Czechia?

The EU’s newly approved Artificial Intelligence (AI) Act—coming into force today, April 1—aims to enhance privacy protection and create legal certainty for businesses for all member states. However, Czech experts and industry leaders warn that the regulation could stifle innovation and drive investment away from the region.

The new regulation, which introduces stricter regulations on artificial intelligence applications, bans certain high-risk uses such as real-time facial recognition and automated CV screening for job applications. It classifies AI systems into four categories based on their risk level, focusing particularly on high-risk and prohibited AI systems.

It bans AI systems that manipulate or deceive individuals, exploit vulnerabilities, or discriminate unfairly. It bans systems that assess criminal risk based on personality traits, develop databases without consent, infer emotions in certain settings, or use biometric data to deduce sensitive information.

It also restricts real-time remote biometric identification in public spaces for law enforcement, except in specific high-risk situations. 

If companies don’t comply with the prohibition of these AI practices, they could be fined up to EUR 35 million (CZK 874 million) or up to 7 percent of their total worldwide annual turnover.

Supporters argue the law will safeguard citizens’ rights. “We do not want a social scoring system like in China, and the AI Act should help prevent that,” Petra Stupková from the Czech Association of Artificial Intelligence told Czech media outlet iDnes.cz. However, she acknowledged that some restrictions might need adjustments to accommodate research and innovation.

Despite these assurances, critics argue that the AI Act adds another layer of bureaucracy, making Europe a less attractive destination for AI investment. Last year, U.S. companies invested around USD 100 billion (CZK 2.3 trillion) in AI development—10 times more than Europe. Countries in the Middle East and China are also outpacing the EU in AI investment and adoption.

“The EU lacks its own foundational AI models, social networks, and distribution channels,” said Jan Romportl, an artificial intelligence expert. “We are already falling behind, and another regulation will only worsen the situation.”

A key concern is the potential for exemptions that could allow restricted AI applications, such as facial recognition in specific scenarios. For instance, Prague’s Václav Havel Airport already employs AI-powered security cameras, and lawmakers are debating their use in football stadiums to enhance safety. This could lead to easy misuse, some warn.

Czech senior member of parliament Patrik Nacher supports the idea but warns against broader implementation. “I wouldn’t want this to become a slippery slope, where facial recognition expands from stadiums to nightclubs and eventually to restaurants,” he said.

Beyond startups, larger technology firms may also reconsider their presence in the European and Czech market due to strict regulatory compliance costs. The AI Act includes significant penalties, calculated as a percentage of a company’s total turnover.

“One day, even tech giants like Meta might decide enough is enough,” warned Zdeněk Valut, director of YDEAL Group. Patrik Tovaryš of Meta echoed this sentiment, arguing that regulation has often hindered, rather than encouraged, innovation in Europe.

With Western firms potentially scaling back, experts fear that Chinese companies could step in, increasing their influence over European AI infrastructure. “China often offers investment in digital governance in exchange for using its AI tools,” Valut explained. “That poses a real risk of European data falling into the hands of foreign governments.”