Creative Commons' Q2 Policy Review

Our Q1 policy review highlighted Creative Commons’ work supporting copyright exceptions for text and data mining (TDM), as well as ways for rightsholders to express their preferences about the use of their works in TDM, including in AI training. In Q2, this continued to be a core focus, as the EU sought to finalize its Code of Practice for general-purpose AI developers in light of the AI Act, which has a strong focus on copyright and transparency-related issues, seeking to clarify how developers should respect “opt-outs.” With the kickoff of the public feedback phase of CC signals, we’re excited to advance concrete ways that opt-outs may be effectuated while also allowing people to broadly opt-in to particular uses.

Policy at EU level

Direct engagement with (and influence on) implementation of the EU’s AI act: We have been actively inputting into the AI Office’s work on implementation of the AI Act, via the Code of Practice, in particular in relation to copyright (defending TDM exemption for AI training and its opt-out regime) and transparency (seeking to balance genuine transparency improvements against overly bureaucratic, administrative burdens). The overarching objective: Ensure exceptions and limitations as enshrined in EU copyright law are not undermined or called into question in the letter or spirit of the EU’s AI legislation. 

Informal community/coalition building: Through direct engagement in Brussels, we continue to build and extend informal collaboration with like-minded organisations (e.g., Open Source Initiative, Wikimedia Europe, Wikimedia France, Open Forum Europe, Open Future, Internet Archive, C4C, Free Software Foundation, KR21, Communia, etc.) creating opportunities to be associated with broader regional policy activities. 

Formal engagement with EU policymakers: In Q2, we have promoted open sharing, open science, and open culture through direct engagement with EU policymakers (European Commission cabinets and services; MEPs and Member States) on policy developments that impact our community. We also attended an invite-only European Commission workshop on Generative AI and creativity, where CC presented our views, amongst other things, on preference signals.

Coalition building with ‘like-minded’ peer organisations: We also continue to leverage our relationships, such as our elected Board-member position at the European Internet Forum, seeking to ensure non-commercial, public-interest, community-driven organisations’ views and input are considered in the EU’s policy-making processes. For example, we attended the European Internet Forum General Assembly as well as its Governing Board and Steering Committee meetings. The EIF is a multistakeholder forum based on neutrality, diversity of views, and impartiality.  

Policy at the UN and World Intellectual Property Organization

International treaties and agreements shape national copyright laws. Creative Commons has long been part of a broad, diverse coalition of voices (the Access-to-Knowledge, or A2K, Coalition) working to ensure that the public interest is well-represented in key fora.

For instance, in Q2, we participated in the 46th session of the Standing Committee on Copyright and Related Rights of the World Intellectual Property Organization (WIPO), which is a specialized agency of the United Nations (UN). CC advocated for clearer and mandatory copyright exceptions to support libraries and archives, and raised concerns about how a proposed Broadcasting Treaty could encroach on the public domain, curtail existing limitations and exceptions, and prevent legitimate uses of openly licensed content. You can watch our two short statements: https://webcast.wipo.int/video/SCCR_46_2025-04-08_PM_124817?startTime=3842 and https://webcast.wipo.int/video/SCCR_46_2025-04-09_AM_124821?startTime=4643. 

In addition, we responded to the UN’s Office of the High Commissioner on Human Rights (OHCHR) “Call for contributions on artificial intelligence and creativity.” Our contribution explored both the concerns around generative AI today and the wide array of creators and creativity that uses AI in generative and other forms. With that in mind, we offered the UN thoughts on how narrowly tailored copyright policy can support thriving creativity, including using generative AI.

We also spoke at the WIPO Conversation on AI and IP about the essential role of CC's copyright licenses and tools, which constitute the open infrastructure that underpins the commons. Given the major disruptions that AI technologies have brought about, many rightsholders are rethinking how they want to share in the commons and whether the infrastructure responds to their needs. This raises deep uncertainties about the future of the commons and of open sharing in the age of AI. That is why we are developing a new piece of the open infrastructure: CC signals, designed to offer new choices for stewards of content collections to signal preferences when sharing their works. This framework is a critical element supporting the infrastructure for open sharing, ethical AI development, and open governance models to empower creators and rightsholders all around the world. You can watch our short intervention: https://webcast.wipo.int/video/WIPO_IP_CONV_GE_25_2025-04-24_PM_124880?startTime=10075.

Data Governance and CC Signals’ Policy Implications

We also continue to work with broad coalitions to support effective data governance for cultural works and institutions. In February, we joined 40 organizations and networks in a letter encouraging the European Union to increase funding for a dedicated “data space for cultural heritage” – infrastructure for cultural institutions to make available digitized objects from cultural institutions. We followed up by participating in the Europeana 2025 conference in Warsaw, Poland. We spoke on the panel ‘Opening up the data space: expanding access and reuse’ and addressed the key questions at the heart of making the common European data space for cultural heritage a place for broad and responsible heritage data sharing and reuse across sectors, especially in the context of AI. Preserving, protecting, and reusing our shared heritage are essential ingredients to really bring it to life in the digital space in fresh and exciting ways and to ensure its sustainable transmission to future generations.

We spoke at the Annual Conference of the Luxembourg Association of Archivists (VLA) whose theme this year was “AI - Chances and Opportunities.” This event was a great opportunity to learn more about the immense potential of AI to assist us in understanding our past based on archives’ rich heritage - if we approach it responsibly. We spoke about copyright, open access, the Creative Commons licenses, tools, and signals, AI, and more, all with the aim of supporting archives’ mission to guarantee access to our documented memory. 

Data governance is also key in AI, of course, including transparency around training data. Transparency can serve many different ends – e.g., helping with safety auditing, addressing bias, or ensuring rightsholders have requisite knowledge to take advantage of “opt-outs.” In the EU, our advocacy on the Code of Practice explored how transparency for copyright purposes can be achieved through high-level disclosures of sites or datasets used. At the same time, in California, we joined a coalition in pushing back against AB 412, which would impose impossible and unnecessary requirements for developers to identify specific rightsholders for each individual piece of training data.