This TechieKat is thrilled to attend the Digital Knowledge – The Library and Copyright in a Global Digital Economy conference which is taking place today and tomorrow in Stockholm, Sweden.
As reported here, the event is organized by the National Library of Sweden in collaboration with the national IPO Swedish Intellectual Property Office (PRV) and the Stockholm Centre for Commercial Law at the Law Faculty at Stockholm University.
Here is this TechieKat’s report on the first day of the conference.
Opening and Keynote Speech
The Conference Planning Committee, Jerker Rydén (National Library of Sweden), Sofie Grettve von Rosen (Swedish Intellectual Property Office), and Kate Parson (EODOPEN, National Library of Sweden) welcomed the attendees and provided practical information about the two-day event. Jerker underlined that AI is a trending topic that can be approached from different views, highlighting that the conference addresses it primarily from libraries' points of view.
The event was opened by Karin Grönvall (National Librarian, National Library of Sweden), Anna Jardfelt (Director General, Swedish Intellectual Property Office), and Jan Rosén (Professor, SCCL Stockholm University). Karin provided insights into the National Library of Sweden and emphasized that libraries must prepare for the future, particularly considering AI and cross-border copyright issues. Anna stressed the long tradition of the Swedish IPO, whereas Jan underlined the time spent organizing the event and the historical importance of the venue.
Then, it was time for the keynote speech. Shira Perlmutter (Register of Copyrights and Director, US Copyright Office) explained the main copyright issues, including digital replicas, copyright liability and infringement, as well as the implications for copyright users and libraries from the US perspective. The speaker also provided highlights about the report on digital replicas released by the US Copyright Office, the recently introduced federal legislation, and the No Fakes Act, which comprises similar provisions to the recommendations in the report. Shira underlined that in light of US lawsuits on AI issues, it remains to be decided whether and how fair use would apply as court decisions would come out.
Session A - Helicopter Perspective
Professor Daniel Gervais (Law School, Vanderbilt University) moderated the session. Daniel stressed that the topic would be approached from the perspectives of libraries, practitioners, and the academy.
First, it was the turn of Dr. Love Börjeson (Head of KB Lab, National Library of Sweden), who shared some insights into modernization practices by libraries. Then, Jerker Rydén (Senior Legal Adviser, National Library of Sweden) discoursed historical views of technological advancements, the digital transformation and its complexity, highlighting the intersection between copyright, training LLMs on data and chips. Last but not least, Professor Noam Shemtov (Queen Mary University of London) addressed AI from infringement angles by comparing US and EU frameworks, explaining views about fair use and its first factor and TDM established in articles 3 and 4 of the DSM Directive.
Daniel Gervais, Jerker Rydén, Love Börjeson and Noam Shemtov |
Session B - Normative Approaches to Regulating AI and Copyright
Jukka Liedes (Finnish Copyright Society) moderated the session. Jukka outlined that most countries do not have legislation on AI, and how quickly such frameworks will be introduced across the board remains to be seen. Then, Ulrike Till (Director of Frontier Technologies, WIPO) pointed out that AI is revolutionizing the creative industries as the technology has been widely embraced. Ulrike also discussed how WIPO Conversations on IP & Frontier Technologies aim to bridge the gap in the impact of those technologies on IP rights.
Ulrike Till, Emmanuelle du Chalard, Jukka Liedes, Anne le Morvan, Tatsuhiro Ueno and Jane Ginsburg |
Emmanuelle du Chalard (Acting Head, Copyright Unit, DG CONNECT, European Union) delved into the TDM provisions of the DSM Directive, emphasizing the requirement of lawful access to data for their application, stressing that no one-fits-all opt-out technical solution exists across the creative industries and explaining the two categories available (a location-based opt-out at the website level, and item-based opt-out at content level). The speaker also provided an overview of copyright-relevant provisions in the AI Act, asserting that the Code of Practice will be drafted in a collaborative and inclusive process.
Jukka Liedes, Ulrike Till, Emmanuelle du Chalard, Anne le Morvan and Tatsuhiro Ueno |
Then, Professor Tatsuhiro Ueno (Waseda University) took the stage. Tatsuhiro stated that Japan was the first country to introduce a TDM exception. He also explained the justification for its introduction and provided examples of cases in light of article 30-4 of the Japanese Copyright Act. After him, Anne le Morvan (Head, Copyright Unit, Ministry of Culture of France) discussed the “Report 2019 of the High Council for Literary and Artistic Property of the Mission on AI and Culture”, the National Commission on AI and the recent CSPLA missions requested by the French Minister of Culture on transparency and remuneration models.
Jukka Liedes, Ulrike Till, Emmanuelle du Chalard, Tatsuhiro Ueno and Anne le Morvan |
To conclude this session, Professor Jane Ginsburg (Faculty Director, Kernochan Center for Law, Media, and the Arts, Columbia University) elucidated the copyright protection requirements for AI-generated output and analyzed US case law involving AI-generated images. Regarding creative control over AI-assisted works, Professor Ginsburg unfolded the challenges when prompting in the sense of lack of control as the machine may “disobey” or “not understand” the prompt, and the registration of “Rose Enigma” on 21 March 2023 by the US Copyright Office. An example of ABBA imitation was also presented during the session. [Merpel is fan of the band - listen here to the one and only ABBA!].
Session C - Training of AI
Professor Jan Rosén (Stockholm Centre of Commercial Law, Stockholm University) moderated the session. Regarding reproduction rights, Anita Huss Ekerhult (CEO and Secretary General, International Federation of Reproduction Rights Organisations) indicated that IFRRO elaborated, in collaboration with Eleonora Rosati, a toolkit that will be launched in two weeks, addressing, inter alia, licensing options and liability in light of AI.
Mats Lindberg, Catherine Zaller Rowland, Tobias Kempas, Jan Rosén, Anita Huss Ekerhult, Susan Allen and James Bennett |
Catherine Zaller Rowland (Vice President and General Counsel, Copyright Clearance Center) discussed the collective AI license recently launched by CCC, the right of reproduction considering Generative AI and infringement challenges in light of the ongoing legal proceedings in the US. After her, Susan Allen (United States Patent and Trademark Office) stated that AI copyright issues are complex and under review in the US. Likewise, she pointed out the need to harmonize the technical layers so that those are translated into machine-readable and understandable.
Mats Lindberg, Catherine Zaller Rowland, Tobias Kempas, Anita Huss Ekerhult, Jan Rosén, Susan Allen and James Bennett |
Then, Tobias Kempas (Attorney at Law, Advokatfirman Vinge KB) took the stage. He emphasized the importance of balancing the protection of rightsholders (e.g., copyright) and the development and use of AI technologies (e.g., freedom of information). After him, James Bennett (Head of Rightsholder Relations and Content Acquisition, Copyright Licensing Agency, United Kingdom) provided views about licensing, remuneration and the TDM exception from the UK perspective. Last but not least, Mats Lindberg (Swedish Copyright Consultants) emphasized that authors are keen on distributing their works under fair remunerations.
Mats Lindberg, Catherine Zaller Rowland, Tobias Kempas, Jan Rosén, Anita Huss Ekerhult, Susan Allen and James Bennett |
Session D - Output of AI
Our very own PermKat, Professor Eleonora Rosati (Law Faculty, Stockholm University), moderated this session, which addressed the protectability of output from legislative/historical and contractual overviews, recent developments, liability in connection with output, and applicable defenses.
Professor Tito Rendas (Executive Dean at Católica Global School of Law, University of Lisbon) underlined how the global map currently looks when it comes to the treatment of copyright subsistence in AI-generated outputs. Then, Professor Alessandro Cogo (Department of Law, University of Turin) and Dr Paweł Kamocki (Legal Expert CLARIN ERIC / IDS Mannheim) discussed their views about the Italian and French legislative proposals regarding the protectability of AI-generated outputs. After them, from a contractual perspective, Matt Hervey discoursed the terms of services when the outputs are generated using third-party AI tools and the potential liability (to be) allocated between tool providers and users.
Last but not least, Molly Stech (General Counsel, the International Association of Scientific, Technical and Medical Publishers) provided insights into US case law in light of commentators’ suggestions that fair use might be broadly applicable in a generative AI context, whether at the input/training stage or output generation.
To close the first day of the conference nicely, a reception at Stockholm City Hall was planned!
Stay tuned for the second-day report from beautiful Stockholm!
IPKat members and friend, Antonios Baris, Eleonora Rosati, Verónica Rodríguez Arguijo and Spyridon Sipetas |
All images (except the last one) by Verónica Rodríguez Arguijo, taken during the event.
[Event Report] #1 Digital Knowledge – The Library and Copyright in a Global Digital Economy
Reviewed by Verónica Rodríguez Arguijo
on
Thursday, September 12, 2024
Rating:
No comments:
All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.
It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.
Learn more here: http://ipkitten.blogspot.com/p/want-to-complain.html