Recently, the European Council has unanimously approved a pioneering law known as the Artificial Intelligence Act, marking a significant step towards harmonising regulations on AI technology across Europe. This landmark legislation introduces a 'risk-based' approach, imposing stricter rules on AI systems deemed to pose higher risks to societal harm, setting a global precedent for AI regulation.

Key Highlights of the Artificial Intelligence Act:

1. Promotion of Safe and Trustworthy AI: The act aims to stimulate the development and adoption of safe and trustworthy AI systems within the EU's single market, balancing innovation with fundamental rights protection.
2. Categorisation and Prohibition: AI systems are categorised based on risk levels. High-risk AI systems, such as those for cognitive manipulation and social scoring, face stringent requirements or outright prohibition. Predictive policing and biometric categorisation based on sensitive attributes like race or religion are explicitly banned.
3. Governance Structure: To ensure effective enforcement, the legislation establishes:
   
   • An AI Office to oversee compliance across the EU.
   • A scientific panel to provide independent expertise.
   • An AI Board comprising member states' representatives to advise on consistent application.
   • An advisory forum for stakeholder input.
4. Penalties: Non-compliance could lead to fines based on a percentage of global annual turnover, ensuring accountability across all scales of enterprises, with proportional fines for SMEs and startups.
5. Transparency and Fundamental Rights: Enhanced transparency requirements ensure accountability in AI deployment, including impact assessments for high-risk systems and public registration of certain AI technologies.
6. Support for Innovation: Provisions for AI regulatory sandboxes facilitate innovation by providing controlled environments for testing and validation of new AI systems.