Automated Journalism
Let's Know Things
Colin Wright
4.8 • 593 Ratings
🗓️ 25 July 2023
⏱️ 22 minutes
🧾️ Download transcript
Summary
This week we talk about Genesis, the Associated Press, and Glorbo.
We also discuss Wikipedia, G/O Media, and ChatGPT.
Show notes/transcript: letsknowthings.com
This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit letsknowthings.substack.com/subscribe
Transcript
Click on a timestamp to play from that location
| 0:00.0 | In mid-July |
| 0:03.0 | In mid-July of 2023, a gaming news site called Z-League published an article about a new feature that would soon come to the remarkably still popular after all these years massively multiplayer online game, World of Warcraft, that feature called Glorbo. |
| 0:33.6 | Among other sub-features, Glorbo would include a new mandatory item called ClickClack, |
| 0:39.3 | a huge new quest to depose Quakian, the aspect of ducks, and a, quote, small, cozy island, |
| 0:46.3 | end quote, called Zoop. |
| 0:48.3 | The publication of this article was celebrated on a World of Warcraft subreddit, a game-specific forum on Reddit, |
| 0:57.5 | not because everyone was super excited about this seemingly bulky new game update, but because |
| 1:02.4 | the Glorbo update was not real. This subreddit community had made it all up. They did this |
| 1:09.3 | because it was becoming increasingly clear that Z-League |
| 1:12.4 | was using some kind of bot to scan the conversations taking place in that subreddit, |
| 1:18.2 | and then using AI to generate articles based on those conversations before then publishing |
| 1:24.0 | those pieces to their site. So rather than reporting on actual news or doing any |
| 1:29.2 | kind of reporting at all, this website was scraping what happened in this popular subreddit, |
| 1:35.3 | grabbing keywords and popular discussion topics, then auto-generating articles based on all that |
| 1:42.1 | hubbub, all that conversation, creating something a bit like game |
| 1:46.3 | industry reporting, but without the traditional trappings of journalism. |
| 1:50.8 | No speaking to the people involved, no fact checking, just sucking up and spitting back |
| 1:56.5 | out the content of a forum, without the permission of those participating in that forum, or compensating |
| 2:03.2 | them in any way. |
| 2:05.3 | This did not go over well with the denizens of this subreddit, hence their effort to intentionally |
| 2:10.7 | feed this seeming AI system, bad data, to poison the well, essentially, making up the concept of glorbo and an increasingly |
| 2:20.4 | elaborate and silly feature set in order to show how ridiculous this approach to pseudojournalism is, |
... |
Please login to see the full transcript.
Disclaimer: The podcast and artwork embedded on this page are from Colin Wright, and are the property of its owner and not affiliated with or endorsed by Tapesearch.
Generated transcripts are the property of Colin Wright and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.
Copyright © Tapesearch 2026.

