Emmanuelle DU CHALARD

What Emmanuelle DU CHALARD has said (3)

• “Thank you to the jury committee for organizing this workshop and for the thanks to the expert and the representative of the creative sectors for sharing their analysis and their views, for proposing ideas, options and recommendation. This idea really contribute to assess this very complex topic of copyright and AI under different perspective. And we find it very valuable. So I will start just by reminding the copyright framework that we have in the EU. As you know, it is based on a number of exclusive rights which allow the right holders to authorise or prohibit the use of their works. And we have also, in the EU acquis a number of copyright exceptions to facilitate certain types of uses of protected works. The text and data mining exception, which has been mentioned this morning, has become increasingly relevant in the context of AI because the training of the models requires access to a vast amount of data, and the TDM techniques are being used in this manner.”
  Promotion of open-source softwares · Intellectual property rights (IPR)
• “I will, thanks. The relevance of this exception has also been acknowledged by the co-legislators in the context of the AI act. The specific feature of this exception is the opt out. It has also been mentioned by many speakers this morning. It allows to reserve the right and then to exclude the application of the exception in order to license the use or to prevent their use for TDM. We are aware of the challenges which are related to the practical implementation of this exception and in particular of the opt out, but we are also aware of the uptake of opt out solution by rights holders in different sectors. And for example, the study recently published by the EU IPO, illustrates very well the various approaches which have been taken by the creative sectors to reserve the right, including some technical solutions which are developed or under development. We are also supporting this practical implementation of the opt out. We have launched already the study to assess the feasibility, but also the opportunity of having a registry of TDM opt out, which can give visibility to the reservation of rights but also help to give information on the licensing condition. And in addition, we would like to further discuss with all the relevant stakeholders how to identify and agree on the opt out solution that needs to be respected by the AI providers. The act is also, of course, very relevant. It includes two copyright related provisions. And as you know, the discussion on the implementation of this obligation is currently ongoing, in particular in the context of the Code of Practice, which aims to make operational the obligation to put in place a policy to comply with EU copyright rules. The Commission is also working to finalize the template for the summary of training data.”
  Intellectual property rights (IPR) · Transparency and oversight of AI-generated content
• “We understand the importance of transparency on the training sources for the enforcement of copyright, and this is the objective of this summary to facilitate rightholders in the exercise and the enforcement of of the right, but also taking into account the need to respect the providers trade secrets. Now, in terms of next steps, as you know, we are approaching the review of the GSM directive, and for this purpose, we will soon launch a study to support the assessment of the GSM directive rules, including the TDM exception. We will also look at the exception provided in article three of the directive, which is the exception of TDM for scientific research purposes and the interplay with the broader exception in article four. We will analyse the benefit brought by the new rules, as well as the implementation issues and the remaining challenges. We also need to look beyond the TDM process and have a broader and comprehensive view and perspective on generative AI. We will consider the different phases of deployment of generative AI and their copyright implication. Based on this analysis, we will consider the need for new measures to support the development of a licensing market that can, on the one hand, facilitate access to high quality data for the development of AI in Europe and, on the other hand, ensure appropriate remuneration for the creators. So we will look at whether new measures are needed to improve the Rightsholders ability to enforce their rights, to facilitate the conclusion of licensing deals based on fair terms, and support licensing at scale. And in doing so, we will take into account the existing licensing practices in the different sectors, the different types of content and the different types of AI application. Thank you.”
  Intellectual property rights (IPR) · Transparency and oversight of AI-generated content