The event brings together the perspectives of computer science, statistics and law. By taking a closer look at AI training and Accountable AI from an interdisciplinary perspective, the workshop aims at providing valuable insights for all disciplines involved. While highly topical research from computer science and statistics will be presented in the workshop (e.g. on training data, biases and memorization) on the one hand, the impact of these results on legal questions on copyright and AI, which are currently subject to the first court decisions in the EU, will be a particular focus on the other hand.
Topics and speakers include:
Fundamental Steps of LLM Training – Matthias Aßenmacher (LMU); AI training from an EU copyright perspective– Lucie Antoine (LMU); Memorization and Data Leakage in Large-Scale Learning: Theoretical Limits and Practical Risks – Roi Livni (TAU); Human Perspectives in Machine Learning: Ethical and Epistemic Challenges – Helen Alber (LMU); Authorship and Copyright Challenges in AI-Generated Art: A Value Chain Perspective – Shahar Sarfaty (TAU); Not All Similarities Are Created Equal: Leveraging Data-Driven Biases to Inform GenAI Copyright Disputes – Amit Bermano (TAU); Majoritarian Signals: Using Generative AI to Guide Judicial Decisionmaking – Uri Hacohen (TAU).
For more information and registration please contact Dr. Lucie Antoine (lucie.antoine@jura.uni-muenchen.de).