Deep learning models struggle to forget data due to their structure and size, posing a challenge to the Right to be Forgotten.