Data trawling to train AI models is the next big thing to solve

Description

When Large Language Models (LLMs) such as ChatGPT are trained, they often use publicly accessible information on the Internet as one of their sources. But the Internet contains personal data. How can these data be protected, especially since it is very unclear what these AI models do with them? Law & technology scholar Taner Kuru tried to find a solution within the EU’s General Data Protection Regulation (GDPR).