AI Chatbots Spreading Russian Misinformation: NewsGuard Report
The rise of AI-powered chatbots has brought about a concerning trend, as leading chatbots are found to be regurgitating Russian misinformation, according to a recent report by NewsGuard shared exclusively with Axios.
In a study conducted by NewsGuard, 57 prompts were entered into 10 leading chatbots, including OpenAI’s ChatGPT-4, You.com, Grok, Inflection, Mistral, Microsoft’s Copilot, Meta AI, Anthropic’s Claude, Google Gemini, and Perplexity. Shockingly, the chatbots spread Russian disinformation narratives 32% of the time, often citing fake local news sites created by American fugitive John Mark Dougan as a reliable source.
NewsGuard co-CEO Steven Brill expressed alarm at the findings, emphasizing the need for special attention to news and information in the industry. He advised users not to trust answers provided by most chatbots, especially on controversial issues related to news.
The implications of this trend are significant, particularly in a year that will see the U.S. presidential election and numerous other elections worldwide. Sen. Mark Warner raised concerns about the rise in misinformation efforts, noting that Americans are increasingly susceptible to believing conspiracy theories.
Furthermore, covert influence campaigns are utilizing chatbots to aid in their efforts, as highlighted in a recent report by OpenAI. Despite efforts by leading AI companies to curb the spread of misinformation, Warner expressed disappointment in the lack of substantial progress.
In a separate development, NewsGuard found itself under scrutiny from House Oversight Committee Chair James Comer, who launched an investigation into the organization. NewsGuard refuted claims of serving as a non-transparent agent of censorship campaigns, emphasizing its commitment to defending First Amendment rights as a journalism organization.
The intersection of AI technology, misinformation, and the upcoming elections underscores the critical need for vigilance and discernment in consuming information from chatbots and other sources. As users increasingly rely on AI for information, ensuring the accuracy and reliability of the content presented is paramount in safeguarding against the spread of disinformation.