Member States should use AI to improve access to information and knowledge, and ensure AI actors respect freedom of expression and access to information in automated content generation, moderation and curation. Frameworks, including regulation, should make online information operators transparent, give users a diversity of viewpoints, notify users promptly of reasons for content removal or other treatment, and provide appeal and redress. Member States should invest in digital, media and information literacy to counter disinformation and hate speech, including understanding recommender systems, and enable media to report on AI and use it ethically.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.