Fans of Blizzard’s iconic MMORPG World of Warcraft are rebelling against trite AI news websites that scrape the game’s subreddit for content.
Users of the MMO’s popular subreddit noticed that gaming news website ZLeague was scraping the forum for news content. Posts that were popular on the 2.4 million user strong subreddit would quickly end up in AI-written articles.
In an attempt to combat the aggressive AI, WoW fans created myriad fake articles to trick the bot into publishing a number of fake news stories. Out of all of the posts, one stuck with fans the most: the introduction of Glorbo.
Posted by Reddit user Kaefer_Kriegerin, a post titled “I’m so excited they finally introduced Glorbo” was scraped by ZLeague’s AI and turned into a news article. The obvious troll post claimed that the character had been teased in “Hearthstone since 1994”. Even funnier, the post included a call out to news-writing bots that was published in ZLeague’s piece.
“Honestly, this feature makes me so happy,” Kriegerin wrote. “I just really want some major bot operated news websites to publish an article about this.”
In the article, archived here, ZLeague’s moronic AI writer included the quote in full, exposing itself even further as a bot-operated publisher. Following the article’s publication, all WoW-themed posts have been removed from the website, but ZLeague is still publishing content.
AI-written content is becoming a major issue on the internet as websites like MSN publish troves of misinformed articles. With dataset biases in AI tools such as ChatGPT parroting alt-right conspiracy theories including holocaust denial, the current state of artificial intelligence is not smart or responsible enough to take over newsrooms.
Nevertheless, tech giants such as Google are still trying to force the technology into every aspect of internet consumption. In recent weeks, Google has even begun pitching an AI journalist that will replace human writers to major publications such as The New York Times.
At the time of writing, there is little AI regulation. While the United States courts have decided that AI-generated content cannot hold up to copyright laws, and Japan has banned AI art for commercial use, there is still major leeway for companies to use and abuse the technology.