An AI-powered gaming news content mill was recently pranked by Redditors with a fictitious character named “Glorbo” in the popular game World of Warcraft. By excitedly discussing a game character they invented themselves, pranksters tricked an AI system into writing an article about their character.
Ars Technica reports that Redditors recently used a fake character from the well-known game World of Warcraft named “Glorbo” to trick an AI-powered gaming news content mill into publishing an article about the fake character.
In a interesting battle of human ingenuity versus artificial intelligence, a Reddit user known as kaefer_kriegerin crafted a fake announcement about the introduction of a non-existent character, “Glorbo,” into the World of Warcraft game. This ruse was designed to expose the automated content scraping practices of a gaming news content mill known as “the Portal” which operated by Z League, a company known for offering cash prizes for gaming tournaments.
The prank was successful, with the Portal publishing an article about the imaginary character, Glorbo. The article, presumably written by an AI under the pseudonym “Lucy Reed,” not only regurgitated the Reddit post but also added nonsensical details. The incident highlighted the lack of human oversight in the content generation process of websites that use AI to create content.
The Reddit user kaefer_kriegerin’s original post read, “I’m so excited they finally introduced Glorbo!!!” The post was intentionally vague about what Glorbo was supposed to be, and it contained factual errors that a human reader would likely catch. For instance, it referenced Hearthstone, another game, as being released in 1994, even though the game was published in 2014.
The prank quickly gained traction among other Reddit users, who joined in to enhance the algorithmic profile of the Glorbo post, making it more attractive for bot harvesting. One commenter even announced a fake AMA (Ask Me Anything) session with the non-existent dev team behind Glorbo.
As the news of the Glorbo prank began to spread rapidly on social media, The Portal took down its post on Glorbo and reportedly removed all World of Warcraft content from its site. At least one other website published a Glorbo article which was also deleted.
The Glorbo example proves that for all the hype AI systems like OpenAI’s ChatGPT are receiving, humans are still far better than machines at spotting a good old fashioned prank.
Read more at Ars Technica here.
Lucas Nolan is a reporter for Breitbart News covering issues of free speech and online censorship. Follow him on Twitter @LucasNolan