Computer Science Meets Law 3rd Annual Workshop: Generative AI and Intellectual Property

We were delighted to host an insightful and groundbreaking workshop co-organized by LMU Munich and TAU The Chief Justice Shamgar Center for Digital Law and Innovation. This full-day event brought together experts from computer science, law, and intellectual property (IP) to address the complex challenges posed by Generative AI (GenAI) and its impact on copyright law, as well as exploring potential legal frameworks that could guide the future of intellectual property in the age of AI.
The workshop offered a comprehensive and interdisciplinary exploration of a wide range of issues surrounding GenAI and its intersection with intellectual property:
1. GenAI & Copyright Challenges: (Moderated by Prof. Niva Elkin-Koren (TAU))
• CS: Shahar Sarfaty (TAU) presented on the GenAI Value Chain and Copyrightability, focusing on the common practice of text-to-image generation. He explored the legal implications of such AI-generated content, proposing a quantitative approach to address the complexities of multiple contributors in a single GenAI creation.
• Law: Prof. Matthias Leistner (LMU) discussed the challenges of applying European copyright law to AI-generated works. He focused on practical issues such as proving authorship and the transparency of AI creation, raising questions about how current laws can be adapted to protect new types of digital content.
• CS: Prof. Roi Livni (TAU) presented on Copyright & Privacy, questioning whether copyright law could be reduced to privacy concerns. He highlighted the potential of using differential privacy in AI models to ensure privacy protection, presenting it as a more conservative approach than traditional copyright laws.
• Law: Prof. Michael Birnhack (TAU) followed up by discussing the Overlap Between Copyright and Privacy, outlining the similarities and differences between these two legal concepts and their relevance in both technology and law.
2. IP, Technology, and Economic Implications (Moderated by Dr. Uri Hacohen (TAU)):
• Law: Prof. Michal Gal (Haifa University) explored the Theory of Trade Secrets & Fencing Costs, addressing protection on big data, focusing on the paradox of trade secret law, which requires companies to make significant investments in protecting their secrets while trying to limit unnecessary fencing costs.
• Economics: Dr. William Lehr (MIT) and Dr. Volker Stocker (Weizenbaum Institute) examined GenAI & Creators, addressing the ongoing evolution of technology and its implications for creators. They discussed the challenges current legal regimes face in light of technological innovation and how optimal fencing costs might change depending on the context.
3. Can GenAI Inform Legal Standards? Challenges and Opportunities (Moderated by Prof. Assaf Hamadani (TAU)):
• CS: Prof. Shay Moran (Technion), in collaboration with other experts, presented on Credit Attribution, discussing the concept of differential privacy and its relevance for AI-generated content. He focused on defining credit attribution in GenAI, and why it's essential to ensure proper recognition of contributors.
• CS: Prof. Amit Beremano (TAU) introduced a method for Quantifying Originality in Stable Diffusion, presenting an innovative approach to assessing originality by measuring tokens used by models in generating content.
• CS: Prof. Gal Chechik (Bar Ilan) discussed the challenges in Visual GenAI, highlighting issues such as data distribution imbalance and the need for AI systems to adapt to the complexities of working with diverse types of data in AI-generated images.
• Law: Prof. Niva Elkin-Koren & Dr. Uri Hachoen (TAU) presented on Majoritarian Bias, proposing the use of AI models to understand judicial standards and benchmarks, drawing on their research in the context of decision-making processes.
• Law: Prof. Talia Fisher (TAU) delved into GenAI and Evidence Law, discussing how AI tools could revolutionize the way evidence is processed, highlighting both the benefits and potential risks of algorithmic adjudication.
• Law: Prof. Orna Rabinovich Einy and Prof. Avital Mentovich (Haifa University) focused on GenAI and Legal Procedures, examining the role of large language models (LLMs) in enhancing access to justice, improving procedural fairness, and transforming alternative dispute resolution.
Keynote Session:
The workshop concluded with an inspiring keynote by Prof. Matthias Leistner (LMU), who delivered a presentation on “Artificial Intelligence as a Global Challenge for Copyrights: A Comparative Perspective.” Prof. Leistner explored the evolving landscape of copyright law in the EU, breaking down the different types of rights—such as moral rights, personality rights, trademarks, and commercial copyrights—and their applicability to AI-generated content. He also discussed the EU AI Act, analyzing criticisms and its potential as a global standard for AI regulation. Finally, he provided a comparative analysis of copyright laws in the US, China, and Japan, assessing whether EU law could become the central framework for AI regulation worldwide.
Panel Discussion:
The event concluded with a dynamic panel discussion on AI & IP Challenges and Opportunities, moderated by Prof. Niva Elkin-Koren (TAU). The panel featured renowned experts, including Prof. Amir Khoury (TAU), Adv. Asa Kling (Head of Intellectual Property Practice, Naschitz Brandes Amir), and Prof. Matthias Leistner (LMU). The discussion centered around the potential and challenges of integrating AI into IP law, emphasizing the need for a collaborative approach to shaping future policies.