French Council of State Issues Opinion on Draft Law Regulating Use of Copyrighted Works in AI Development

The Conseil d'État has issued an advisory opinion on a draft law aimed at regulating the use of copyrighted works in the development of artificial intelligence systems. The opinion evaluates the legal framework proposed by the French government, particularly in relation to intellectual property rights, exceptions for text and data mining, and the balance between innovation and rights holders’ protections.

The draft legislation seeks to clarify the conditions under which protected content may be used for training AI models, including transparency obligations, potential authorization mechanisms, and safeguards for creators. The Council of State’s analysis highlights key legal considerations, including compatibility with European Union law and the need to ensure proportionality between technological development and copyright protection.

The opinion may lead to adjustments in the legislative proposal before it is formally introduced or adopted. As such, it represents an intermediate step in the policymaking process, with potential implications for AI developers, digital service providers, and rights holders operating in France and across the European Union.

The advisory opinion issued by the Conseil d'État on the draft law regulating the use of copyrighted works in artificial intelligence development represents a medium-impact regulatory development, as it may lead to significant legal and compliance changes, but does not yet impose binding obligations since it remains at a pre-legislative stage.

• Potential Regulatory Changes to the Use of Protected Data: The draft law may introduce new conditions governing the use of copyrighted content for AI training, directly affecting developers and technology providers.
• Impact on Copyright and Text and Data Mining (TDM) Exceptions: The initiative could redefine the balance between innovation and intellectual property protection by adjusting legal exceptions and limitations.
• New Transparency Obligations: AI developers may be required to disclose information about training datasets, increasing compliance and reporting burdens.
• Potential Authorization or Licensing Mechanisms: The framework may introduce requirements for consent or compensation when using protected works, impacting current business models.
• Alignment with European Union Regulation: The opinion considers consistency with EU law, potentially extending its broader regulatory influence beyond France.
• Impact on Digital Ecosystems and Platforms: Technology companies and digital platforms may need to adapt data collection and processing practices.
• Uncertainty Due to Pre-Legislative Stage: As this is an advisory opinion, the final regulatory outcome remains uncertain and subject to legislative developments.
• Legal Rather Than Technical Focus: The initiative does not introduce technical requirements (e.g., RF, EMC, safety), but instead focuses on legal compliance related to data use and copyright.

Who?

• AI Developers: Companies training AI models may face new restrictions or licensing requirements.
• Technology Companies and Digital Platforms: Particularly those handling large volumes of copyrighted content.
• Copyright Holders (creators, publishers, media): May benefit from stronger protections or compensation mechanisms.
• Cloud and Data Infrastructure Providers: Indirectly impacted through changes in demand and compliance requirements.
• Regulators and Policymakers: Responsible for shaping and finalizing the legal framework.
• Enterprise AI Users: Organizations deploying AI may need to ensure compliance with new legal obligations.

Conclusion

The opinion issued by the Conseil d'État constitutes a medium-impact initiative, as it signals potential regulatory changes in the governance of copyrighted data used in AI, without introducing immediate obligations.

The primary impact lies in the possible transformation of the legal framework for AI training, particularly regarding copyright, transparency, and licensing. This could lead to significant adjustments in development practices, operational costs, and business models.

While no immediate action is required, the ongoing legislative process introduces uncertainty and calls for close monitoring by stakeholders. In the medium term, the regulation could significantly influence how AI systems are developed and deployed in France and potentially across the European Union