Spain cracks down on unlabelled AI content with massive fines

Spain's government has approved a bill imposing significant fines on companies that fail to label content generated by artificial intelligence (AI), aiming to tackle the spread of "deepfakes".

The legislation aligns with the European Union's AI Act, which enforces strict transparency requirements for high-risk AI systems, Digital Transformation Minister Oscar Lopez announced.

"AI is a powerful tool that can enhance our lives … or be used to spread misinformation and undermine democracy," Lopez said.

Spain is one of the first EU nations to implement these regulations, which are far stricter than the voluntary compliance framework used in the United States.

Lopez stressed that deepfake threats affect everyone, as AI-generated videos, images, and audio can be falsely presented as real.

The proposed law, pending lower house approval, categorises the failure to label AI-generated content as a "serious offence", punishable by fines of up to €35 million (£30 million) or 7% of a company’s global annual turnover.

Since OpenAI launched ChatGPT in 2022, concerns over AI’s societal impact have surged, making regulation a priority.

The bill also prohibits the use of subliminal AI-driven techniques to manipulate vulnerable groups. Lopez cited chatbots urging gambling addicts to place bets and toys encouraging children to take dangerous actions as examples.

Additionally, AI-based biometric profiling—where individuals are assessed based on personal traits or behaviours—would be banned, except for national security-related real-time biometric surveillance in public spaces.

The newly formed AI regulatory body, AESIA, will oversee enforcement, though specific cases involving data privacy, crime, elections, credit ratings, insurance, or financial markets will be handled by their respective regulatory authorities.