- <span class="news-text_medium">2 February 2025:</span> The prohibitions on AI systems identified as presenting "unacceptable risks" came into force. This includes bans on systems that exploit vulnerabilities in specific groups, use subliminal techniques or are applied for social scoring.
- <span class="news-text_medium">2 May 2025:</span> The AI Office will be responsible for supporting the creation of codes of practice for providers of general-purpose AI models. If these codes are not completed by 2 August 2025 or are deemed insufficient, the European Commission may step in to establish uniform rules for these providers.
- <span class="news-text_medium">2 August 2025:</span> The obligations for providers of general-purpose AI models will take effect, including requirements for technical documentation, compliance with copyright laws and transparency around training data. Additionally, EU Member States will begin enforcing rules regarding the notification of authorities, penalties and administrative fines.
Organisations engaged in AI development or deployment within the EU should ensure they are prepared to meet these deadlines and comply with the new regulatory framework.
