People still use Medium? Garbage platform for ages.
AI Slop Is Flooding Medium
21–30 of 76 posts
Re: AI Slop Is Flooding Medium
#22> CEO Tony Stubblebine says it “doesn’t matter” as long as nobody reads it.
It seems like the leaders of all of these online platforms are going to just bury their heads in the sand and pretend that this massive problem isn't one.
This AI trend is absolutely going to destroy many (most?) of these companies. For better or worse, the new age of the internet is already here.
Re: AI Slop Is Flooding Medium
#23Definitely not just Medium. It's everywhere. I have noticed that someone immediately calls out pretty much every bit of content as AI generated, whether it obviously is or not. I guess this is a best case scenario. I do wonder what will spring up next to allow high quality content to be found.
That sounds exactly like what an AI would say!
Re: AI Slop Is Flooding Medium
#24Previously, spam has been reported and filtered via "flag" and "report" buttons. Why is that not sufficient for site operators to separate "AI Slop" from other articles? If existing sites choose to tolerate "AI Slop" (why doesn't HN have this problem?), does that create a market opportunity for competitors with better filters?
To be honest, I do use GenAI to help me format my comments, including this one. However, I put in substantial input by outlining my thoughts in detail, so I treat it more as a sophisticated grammar checker. As a non-native speaker, my writing is sufficient for business, but I don’t consider myself a strong writer. GenAI helps me convey my thoughts more clearly and efficiently, which has been tremendously valuable.
Re: AI Slop Is Flooding Medium
#25one of the saddest parts of the whole thing is finding out how many fucking arseholes out there have been waiting to flood the internet with crap, and had previously only failed because they were too lazy to do all that typing. just absolutely embarrassing for us all.
It's an incentives problem, and the ad industry is largely to blame. It's the same cause as much of the destruction and centralization of the web, the rise of clickbait, rising extremism and political polarization, negativity-driven engagement, and myriad other public problems. An attention-drivin economy is disastrous for mental and public social health. And some people still defend the ad industry, even going so far to laud it as a public good, even talking about it as if it's a charity and not a very rich industry, funding a "free Internet".
We missed the boat long ago on dismantling or neutering the ad industry before it did massive irreparable damage. Whatever we do to fix this mess is still going to leave prominent scars for generations.
Re: AI Slop Is Flooding Medium
#26> CEO Tony Stubblebine says it “doesn’t matter” as long as nobody reads it. It seems like the leaders of all of these online platforms are going to just bury their heads in the sand and pretend that this massive problem isn't one. This AI trend is absolutely going to destroy many (most?) of these companies. For better or worse, the new age of the internet is already here.
is there anything they can realistically do?
Re: AI Slop Is Flooding Medium
#27> CEO Tony Stubblebine says it “doesn’t matter” as long as nobody reads it. It seems like the leaders of all of these online platforms are going to just bury their heads in the sand and pretend that this massive problem isn't one. This AI trend is absolutely going to destroy many (most?) of these companies. For better or worse, the new age of the internet is already here.
Why wouldn't they? They've spent the last decade pretending that the clickbots sustaining the internet ad model are no big deal and apparently the market agrees with them!
Re: AI Slop Is Flooding Medium
#28Definitely not just Medium. It's everywhere. I have noticed that someone immediately calls out pretty much every bit of content as AI generated, whether it obviously is or not. I guess this is a best case scenario. I do wonder what will spring up next to allow high quality content to be found.
There's so many LLM bots on Twitter now, presumably trying to game the revenue sharing system. Some of them even seem to be getting fancy with multi-modal models because they try to respond to the content of images rather than just the text like they used to do.
Re: AI Slop Is Flooding Medium
#29Previously, spam has been reported and filtered via "flag" and "report" buttons. Why is that not sufficient for site operators to separate "AI Slop" from other articles? If existing sites choose to tolerate "AI Slop" (why doesn't HN have this problem?), does that create a market opportunity for competitors with better filters?
A lot of the time, AI slop is difficult to discern from authentic content until you’re somewhere in the middle, at which point something starts to feel off about it. At least, that’s been my experience coming across AI-written Blender tutorials on YouTube. All this makes it difficult to quickly report as spam. There’s also a lot of plausible deniability since a lot of people have developed a hair-trigger response to…
"A lot of the time, AI slop is difficult to discern from authentic content until you’re somewhere in the middle, at which point something starts to feel off about it."
And it's only going to get worse. I don't know what the limits of LLMs are, even before we start discussing future architectures and capabilities, but it certainly seems to me that if they just scale a bit farther on their current trajectory it is going to be very hard to detect AIs simply by the content.
One of the adaptations I've adopted is that if your content looks like it was generated from a simple prompt, no matter how long the resulting content, I just toss it. "Write me a beginner's tutorial for Python" isn't going to produce a document that the internet needs another copy of or that social media sites need to spend time linking to. Unfortunately that means even if the "tutorial" was human-generated, it's probably getting the axe too, unless it has a very interesting or unusual approach, but it isn't clear to me that that is actually a large problem.
Re: AI Slop Is Flooding Medium
#30Currently on ChatGPT three of the top picks on the explore page are "AI Humanizers" for making generated content appear like it was created by a human. Slop slop slop.