AI chatbots more easily misled by left-wing claims, study finds

A new study suggests that popular artificial intelligence chatbots may exhibit a bias, frequently falling for misinformation attributed to the political left more often than the right. The research, conducted by the organization Just Facts, examined three of the four leading AI chatbots.

Researchers presented these chatbots with various claims. The study found that the AI systems were more prone to being misled by statements linked to the left-wing political spectrum. This susceptibility was reportedly more pronounced than when presented with claims associated with the right.

The findings raise concerns about the impartiality of AI tools that are increasingly used to process and disseminate information. The study implies that these chatbots may not be as neutral as users assume, potentially skewing the information they provide based on the political leaning of the source material.