Artificial intelligence is a powerful tool, but it is easily exploited to deceive and demoralize the vulnerable. Two common types of AI—predictive and generative—put democracy at risk if they are not properly regulated. Predictive analytics can divide publics into ever more isolated silos, eroding the type of common knowledge necessary for democratic deliberation. Generative AI is making it ever easier to fake images and events. Even when such fabrications are debunked, the air of unreality created by such dissimulation has allowed some politicians to deny real evidence of wrongdoing by claiming evidence of it was “AI-generated.” This dangerous confluence of trends means reformers must work on more than combating misinformation. They also need to resist demoralization, which occurs when citizens start tuning out politics altogether and become too cynical or distracted to engage. AI may also play a small part in addressing the problems of misinformation and demoralization—but real solutions entail a far broader conversation about the future of democracy.