AI Chatbots Fuel Spam Content Across the Internet, Highlighted by Silly Error Messages
-
AI chatbots like ChatGPT are being used to generate spammy, low-quality content across the internet that features error messages when asked to violate policies.
-
Spammy AI content is prolific on sites like Twitter, Amazon product listings, blogs, and more.
-
AI error messages like "I cannot fulfill this request" have become a way for people to detect AI-generated spam.
-
Sites often publish AI content before a human catches issues, allowing silly AI errors through.
-
Experts say better content moderation is needed, but blame should fall more on those exploiting AI for spam rather than the tech itself.