Tech Giant Opts Out of Voluntary Compliance Framework
Meta Platforms Inc. has declined to sign the European Union’s voluntary Code of Practice for General-Purpose AI (GPAI) models, labeling the guidelines as regulatory overreach. The decision, announced by Meta’s Chief Global Affairs Officer Joel Kaplan, underscores growing tensions between Big Tech and European regulators over the governance of artificial intelligence.
Background: The EU’s AI Code of Practice
The European Commission introduced the Code of Practice in June 2024 as a supplementary framework to the EU’s Artificial Intelligence Act. Designed to provide clarity and support for AI developers, the Code outlines best practices for:
- Transparency: Ensuring AI systems are explainable and accountable.
- Copyright Compliance: Addressing intellectual property concerns in AI training data.
- Safety Measures: Mitigating risks associated with high-impact AI models.
While the Code is voluntary, it serves as a precursor to stricter enforcement under the AI Act, which aims to establish harmonized rules for AI deployment across the EU.
Meta’s Stance: Innovation vs. Regulation
In a LinkedIn post, Kaplan argued that the Code introduces “legal uncertainties” and measures that exceed the scope of the AI Act. He warned that the guidelines could stifle innovation, stating:
“Europe is heading down the wrong path on AI. This code could throttle the development of frontier AI models and stunt European businesses.”
Meta’s refusal aligns with a broader industry pushback. Over 40 European companies, including Siemens and Airbus, have urged the European Commission to pause the AI Act’s implementation, citing similar concerns.
Industry Reactions: A Divided Landscape
While Meta has opted out, other tech firms are taking a different approach. OpenAI has pledged to sign the Code, signaling its commitment to aligning with EU regulations. Below is a comparison of key players’ positions:
| Company | Position on EU AI Code | Key Argument |
|---|---|---|
| Meta | Declined to sign | Regulatory overreach could hinder innovation. |
| OpenAI | Committed to signing | Supports harmonized rules for AI safety. |
| Microsoft | Likely to sign | Seeks clarity for compliance under the AI Act. |
Implications for AI Development in Europe
Meta’s decision raises questions about the future of AI innovation in Europe. The company has already paused training its AI models on European data due to GDPR concerns, and its rejection of the Code could further isolate the region from cutting-edge AI advancements.
The European Commission, however, remains steadfast. A spokesperson reiterated the EU’s commitment to “risk-based rules that ensure AI systems are safe and trustworthy,” dismissing calls for delays.
Conclusion: A Balancing Act
Meta’s refusal to sign the EU’s AI Code highlights the delicate balance between fostering innovation and ensuring regulatory compliance. As the AI Act moves toward enforcement, stakeholders will need to navigate these tensions to shape a future where technology and governance coexist.
