Countries with a collective population of four billion will vote for leaders next year. There are fears that recent advances in generative artificial intelligence will make voters more vulnerable to deception than ever. But disinformation has long been a problem, well before the age of deepfakes and large-language models. How worried do we really need to be about AI’s potential to undermine democracy? Chihhao Yu of the Taiwan Information Environment Research Centre explains the threat posed by Chinese misinformation campaigns. We go back to when sensationalist journalism drove America towards war. And Senator Josh Hawley explains why he wants AI to be regulated. John Prideaux hosts with Charlotte Howard and Idrees Kahloon. Runtime: 44 min You can now find every episode of Checks and Balance in one place and sign up to our weekly newsletter. For full access to print, digital and audio editions, as well as exclusive live events, subscribe to The Economist at economist.com/uspod.  To help us to continue to improve our podcasts, we’re inviting listeners to participate in some research, particularly those who haven’t filled in one of our surveys before. We’re looking for people to keep in touch with us via WhatsApp over eight weeks. You can sign up here. Hosted on Acast. See acast.com/privacy for more information.