The article discusses the potential risks posed by AI systems like OpenAI’s ChatGPT in spreading misinformation, particularly in the context of the upcoming 2024 U.S. elections. It highlights the ability of ChatGPT to generate highly coherent and persuasive text on almost any topic, including false or misleading information. Experts warn that bad actors could exploit this technology to create and disseminate disinformation campaigns at an unprecedented scale. The article also mentions the concept of “perplexity,” which measures the uncertainty of an AI model in generating text, with lower perplexity indicating higher confidence and potential for spreading misinformation. Additionally, it explores the idea of an “election hub” to monitor and counter AI-generated misinformation during the election cycle. Overall, the article raises concerns about the potential misuse of AI for spreading misinformation and the need for proactive measures to address this emerging threat.
Source: https://www.businessinsider.com/openai-chatgpt-perplexity-election-hub-ai-information-risks-2024-11