The European Commission has published the General-Purpose AI Code of Practice to complement the European AI Act. Developed by independent experts and extensive stakeholder input, the Code aims to support compliance with upcoming regulations. The AI Act, effective August 1, 2024, enforces compliance on prohibited AI practices by February 2025, notably banning several AI uses like facial recognition and social scoring. From August 2, 2025, general-purpose AI models must document their training data while protecting intellectual property and trade secrets.
"Today's final code of practice for general-purpose AI (GPAI) models still imposes a disproportionate burden on AI providers." - CCIA statement
The General-Purpose AI Code of Practice aims to establish guidelines for compliance with the European AI Act, focusing on safety and security risks associated with AI.
The AI Act bans several uses of AI technologies, including untargeted image scraping for facial recognition and emotion recognition in schools and workplaces.
As of August 2, 2025, general-purpose AI models must report their training data while respecting intellectual property rights and confidential business information.
Collection
[
|
...
]