A scandal surrounding ChatGPT and its allegation of sexual harassment against law professor Jonathan Turley has highlighted the potential pitfalls of using AI chatbots to generate information, as reported by the Washington Post.
As part of a research study, a fellow attorney in California asked ChatGPT to compile a list of attorneys who had sexually harassed someone. Turley’s name was included in the list, and ChatGPT alleged that Turley made sexually suggestive comments and attempted to touch a student during a school trip to Alaska, citing a March 2018 Washington Post article as the source of the information .
However, it turned out that such an article did not exist, there had never been a class trip to Alaska, and Turley denied ever molesting a student.
AI chatbots draw from a vast pool of online content, often pulled from sources like Wikipedia and Reddit, to compile plausible-sounding answers to almost any question. But just because they’re good at predicting which words are likely to go together doesn’t mean the resulting sentences are always true.
Who is responsible when an AI becomes independent?
Kate Crawford, a professor at the University of Southern California at Annenberg and a senior researcher at Microsoft Research, warned The Washington Post of the dangers of AI chatbots: “Because these systems respond so confidently, it’s very tempting to assume they can do anything, and it’s very difficult to distinguish between facts and untruths.”
The incident has also raised questions about who is responsible when AI chatbots generate or disseminate inaccurate information. Eugene Volokh, a law professor at the University of California at Los Angeles, conducted the study in which Turley was named, The Washington Post reports. Volokh told the newspaper that the rising popularity of chatbots is a key reason why there is an urgent need to clarify who is responsible when the AI chatbots spread false information.