Concerns of AI Chatbots Spreading Election Misinformation in the EU
While AI chatbots have been acclaimed for their potential to revolutionize various aspects of our lives, recent research has shed light on their limitations. Researchers from Democracy Reporting International found that AI chatbots are spreading false information about the upcoming European Union elections. The analysis revealed that the chatbots provided incorrect election dates and voting information, as well as broken and irrelevant links. This issue, known as “hallucination,” is a common problem with chatbots. We’ve been keeping a close eye on this issue and conveying our worry since ChatGPT was first released. In our own research as AI LLMs were released to the public, we gathered that these AI systems did not accurately report on voting and political information, and often displayed bias in response to general election related questions.
Developers caution against using chatbots for election-related queries due to their limitations and advise users to seek information from more reliable sources like Google Search. This underscores the necessity for advertisers’ presence in search advertising, ensuring that individuals seeking election information receive accurate details.
AI Chatbots used properly can be a very useful tool, but they do have their shortcomings and the information they provide must be taken with a grain of salt.
