Some writers and media editors applaud the platform’s approach to AI. Eric Pierce, who founded Medium’s biggest pop book Fanfare, says he doesn’t need to block many AI-generated submissions and believes the human moderators of Medium’s boost program help highlight the best of human writing on the platform. “I can’t think of a single piece that I’ve read on Medium in the past few months that even revealed that it was done by AI,” he said. “Increasingly, Medium feels like a place of sanity in the middle of the Internet that wants to be alive.”
However, some writers and editors believe that they are still seeing a lot of AI-powered writing on the platform. Content marketing writer Marcus Musick, who edits several books, wrote a post complaining about how what he suspected to be an AI-generated article went. (Reality Defender analyzed the article in question and estimated it to be 99 percent “likely to be a hoax.”) The story appears to be widely read, with more than 13,500 “clap hands.”
In addition to seeing potential AI content as a reader, Musik believes he encounters it regularly as an editor. He says he rejects about 80 percent of potential contributors a month because he suspects they are using AI. He doesn’t use AI detectors, which he calls “useless,” instead relying on his own judgment.
While the volume of potential AI-generated content on Medium is remarkable, the challenges of moderation the platform faces—how to produce good work and keep the trash out—is one that has plagued the wider web. The AI boom has only exacerbated the problem. While click farms have long been a problem, for example, AI has given SEO-focused entrepreneurs a way to quickly resurrect zombie news feeds by filling them with AI slop. There are all kinds of YouTube hustle entrepreneurs doing get-rich-quick courses that encourage others to build AI slop on platforms like Facebook, Amazon Kindle, and, yes, Medium. (Example headline: “1-Click AI SEO Medium Empire 🤯.”)
“Medium is in the same place as the internet as a whole right now. Because AI content is so fast to generate that it’s everywhere,” said crime consultant Jonathan Bailey. “Spam filters, human moderators, and so on—that’s probably the best tools they have.”
Stubblebine’s argument—that it doesn’t matter if a platform contains a lot of garbage, as long as it effectively promotes good writing and limits access to said garbage—is perhaps more logical than any attempt to completely banish AI slop. His strategy of moderation may be the wisest approach.
It also suggests a future where the Dead Internet theory comes into play. The theory, once the domain of extreme internet conspiracy theorists, argues that most of the internet is devoid of real people and human-generated posts, instead covered with AI-generated slop and bots. As AI productivity tools become more commonplace, platforms that stop trying to turn off bots will incubate an online world where human-made work becomes more difficult to find on AI-filled platforms.
Source link