How much is AI hacking your emotions? (ft.com)

Disinformation and deepfakes are known risks but biased LLMs could also be used to influence public opinion