David Schwartz, the chief technology officer at Ripple, recently called attention to the latest AI controversy on his social media profile.
Schwartz shared a viral Reddit post, which claims that a family suffered from poisoning due to relying on a mushroom identification book that ended up being AI-generated.
The Reddit poster said that the entire family ended up in a hospital after consuming poisonous mushrooms.
The book was allegedly purchased from a major retailer. It contains images of mushrooms that turned out to be AI-generated. Moreover, the book’s text was also likely written with the help of an AI since it even contained some standard replies from a chatbot.
Even though the retailer agreed to refund the book, there are still concerns about other low-quality books that are still available on its website.
The Reddit poster wondered whether or not this sort of negligence could be reported to the police to punish the creators of the book.
“If this report is true, it’s history repeating itself,” Schwartz said in his social media post.
The Ripple executive pointed to Winter v. G.P Putnam’s Sons, a Court of Appeals case from 1991. Two young adults bought a book called “The Encyclopedia of Mushrooms” as a reference guide.
The guide included extensive information on where to find mushrooms, how to collect them and how to cook them. The British book was distributed by G. P. Putnam’s Sons, an American book publisher based in New York. After using the book, the couple became critically ill (to the point where they needed liver transplants).
The couple ended up suing P. Putnam’s Sons for product liability, negligence and false representation. Despite the fact that the mushroom hunters nearly died due to the incorrect information offered by the book, the court ended up siding with the publisher.
The proliferation of low-quality AI-generated books will make it even harder for readers to obtain accurate information.