Open scientific publications and data now constitute a vital resource for training artificial intelligence models. This situation raises new questions for scientific communities: how can they know under what conditions their work is being reused? What guarantees exist for recognition, transparency, and traceability? And how can openness be leveraged to develop AI that is responsible and beneficial to research?
This webinar will address the legal, ethical, and technical challenges associated with the algorithmic reuse of scientific knowledge, with a particular focus on licenses, metadata, training corpora, and mechanisms that make reuse conditions interpretable by automated systems.
Session on 21 September, from 3:00 pm to 4:30 pm
Speakers:
Lionel Maurel, Research Engineer at the Open Data Directorate for Research (CNRS)
Susanna Fiorini, Translator and consultant in translation technologies and multilingual communication (OPERAS)
Pierre-Charles Langlais, Co-founder of PleIAs
The webinar is open to all and requires registration: Inscription Parlons Science Ouverte n°15 – Science ouverte et IA : pour une réciprocité entre producteurs de savoirs scientifiques et concepteurs d’IA | Framaforms.org