- The European Union is assembling leading experts to create the pioneering “Code of Practice” for general-purpose AI models as part of its AI Act.
- This effort underscores the EU’s commitment to setting global standards in AI transparency and risk management.
- “This initiative marks a significant step towards responsible AI governance,” said Nuria Oliver, a distinguished AI researcher.
Discover how the EU is shaping the future of AI with the first “General-Purpose AI Code of Practice” to balance innovation and risk management.
EU’s Groundbreaking Initiative: Drafting the First AI Code of Practice
The European Union is leading a pivotal initiative to develop the first-ever “General-Purpose AI Code of Practice” under its AI Act, aiming to introduce stringent standards for AI transparency, copyright, risk assessment, and internal governance. According to a recent announcement on September 30, this endeavor is managed by the European AI Office and collaboratively supported by a diverse group of nearly 1,000 experts from academia, industry, and civil society. This collective effort seeks to forge a comprehensive framework that addresses the multifaceted challenges posed by AI technologies.
Key Contributors and Roles in AI Governance
The inaugural plenary session, which marked the beginning of this extensive process, featured around 1,000 participants and highlighted four working groups spearheaded by renowned industry leaders, including Nuria Oliver, a prominent AI researcher, and Alexander Peukert, an expert in German copyright law. These groups are tasked with critical areas such as transparency and copyright, risk identification, technical risk mitigation, and internal governance. They will operate between October 2024 and April 2025 to draft the code, engage stakeholders, and refine the framework through ongoing consultations.
Balancing Innovation with Safety: The EU’s Strategic Approach
The AI Act, ratified by the European Parliament in March 2024, embodies a risk-based approach to AI governance. It classifies AI systems into various risk categories, from minimal to unacceptable, with corresponding compliance requirements. This legislation is particularly pertinent to general-purpose AI models, which often fall into higher-risk categories due to their wide-ranging applications and potential societal impacts. Despite some criticisms from major AI firms like Meta, which argue that the regulations could hinder innovation, the EU remains focused on creating a balanced approach that fosters both safety and ethical AI development.
Global Implications and Future Outlook
The EU’s comprehensive and collaborative effort has already yielded over 430 stakeholder submissions, which will play a vital role in shaping the final Code of Practice. By April 2025, this initiative aims to establish a precedent for responsibly developing, deploying, and managing general-purpose AI models, with a strong emphasis on minimizing risks and maximizing societal benefits. As the global AI landscape continues to evolve, this groundbreaking framework is poised to influence AI policies worldwide, particularly as other countries look to the EU for guidance on regulating emerging technologies.
Conclusion
In conclusion, the European Union’s proactive approach in drafting the first “General-Purpose AI Code of Practice” under its AI Act is a significant milestone in AI governance. By bringing together a global coalition of experts and focusing on transparency, risk management, and ethical development, the EU is setting a high standard for AI model governance. This initiative not only aims to safeguard societal interests but also to catalyze responsible innovation in the rapidly advancing field of artificial intelligence, potentially shaping policies worldwide.
