Comment by yoyohello13
18 hours ago
Wow! that comment you linked speaks to me. I was watching a review video for a game yesterday that was so obviously written by AI it was actually painful to listen to. I started reading the comments and there were several people praising this Youtuber on their great use of metaphor. Literally "subbed because you're such an amazing writer." It actually shook me to my core. If those were real people making those comments, there is no hope for us. I need to believe those were mostly bot comments just to maintain my sanity.
At least you didn’t over-react. Some people seem so afraid of consuming AI-generated content that they see it everywhere. A popular YouTuber says “thanks for the pushback” in a YouTube video and has the mob at his door, instantly, despite it not being written by AI. In the comments for any blog post or article that hits the HN front page, there is always someone confidently calling it out for being unmistakeably written by Claude. It’s so tiring. Let’s say the YouTube video that rocked the very foundation of your being was scripted with AI assistance. It was apparently good enough for people to praise it specifically. Why is that a problem? If it wasn’t AI, which may well be the case, would your soul be calmed?
> A popular YouTuber says “thanks for the pushback” in a YouTube video and has the mob at his door, instantly, despite it not being written by AI.
Well that is not really what happened at all. Most people seem to believe Hank Green was honest when he said the line "thanks for the pushback" was in fact made up at the spot and not by an LLM, and the scandal would have died there.
Except in follow-up to that, Hank Green made several very open and personal posts about his LLM usage getting completely out of hand to the extent that his wife and brother called for an intervention about it, that it's been like an addiction to him, and that he had been letting it guide him way too much and how this was a breach of trust to his audience.
There were also one or two other incidents for people who are paying attention that this did another rather large dent in trusting him as a "smart person who says smart things", any more.
What I find tiring: the same. writing. style. everywhere.
Various youtubers I subscribe to are using apparently AI-written scripts now. They all wrote and spoke just fine before, each with their own particular style. Now they just sound like Claude. I know it's still their thoughts, and I'm still interested in what they have to say. Maybe they're using AI to save time, or maybe they've read too much AI text and now that's how they write. But it kinda takes the sparkle and personality out of everything.
What's tiring is being spammed with AI slop every where. A strong counter reaction that may occasionally have false positives is only natural.
>A popular YouTuber says “thanks for the pushback” in a YouTube video and has the mob at his door, instantly, despite it not being written by AI.
In this era, I live by two principles
1. Do my best to avoid AI slop. If you agree, this is straightforward. I don't always succeed and I'm sure I've "fallen for it" in quite a few places, but I disengage if I sense it.
2. Let sleeping dogs lie. Because I'm not some almighty judge of AI, I'm not going around accusing others of doing so. There is zero benefit for me to call out random internet pieces on if something is AI. And ultimately, I'm in discussions to engage with humans, not to monitor bots nor even to argue about the existence of bots (I feel I've said all I can on the topic, I have moved on internally in terms of discussion).
>It was apparently good enough for people to praise it specifically. Why is that a problem?
Writing is subjective, but I can certainly see some dangerous anti-patterns develop if you take bad advice as a beginner, especially advice that ultimately came from hallucinations in an LLM.
I was watching a Youtube video about "developing taste" and how culture has been impacted in the modern age. TYou risk becoming "tasteless" consuming AI slop, because AI by its nature is trying to average out all the inputs and cut of all the edges of any given culture. That's not necessarily a bad thing if you just wanted to get a brief overview of a culture, but is poison if it comes time to try and become part of that culture.
as a huge summation of an interesting video: There's 3 aspects of a culture: embodied,objectified, and institutionalized. it's never been easier to pretend to be a "poser" in this era of AI and be rewarded for it, because AI makes objectified easier than ever. But it doesn't mean it truly embodies you into the culture. And that's dangerous for multiple reasons (in my eyes).
> If those were real people making those comments
Really, does it matter? Lots of YouTube comments are zero-effort attempts to juice a video/channel's engagement metrics (aka "commenting for the algorithm"), and whether they're accomplices/sockpuppets, or just unimaginative parasocial parasites, is basically unknowable from the outside.
LLMs output the average of their training data. On average, people are tasteless and clueless. So it's to be expected that average and especially below-average people find LLM output appealing. Or the tasteless work of an average person.
You realize that when people aren't told an image or text is AI generated, on average they prefer it to human created content, and it's only when they know it's AI that they rate it lower than the human content? You shouldn't be shocked by this, the last 20 years of Hollywood and pop music clearly indicates the the mass market doesn't care about craft or art, it likes to ride the vibes of what's trending.
Yeah, and you might enjoy a meal, and when someone tells you it contained 5 percent human eyeballs to give the sauce that awesome texture, and you know they're not lying, you will puke it out right on the spot.
Then you say "it's because I don't like human eyeballs in my food", and someone and replies "but the mass market clearly doesn't care. You shouldn't be shocked by this."
> when people aren't told an image or text is AI generated, on average they prefer it to human created content
uhm this is so heavily dependent on the kind, type and model of AI used that I'm calling you out on making this "statistic" up on the spot.
but I'm not entirely sure what your point is
if you don't tell people which food has been hyperprocessed and laced with corn syrup or whatever you can imagine, they will also prefer that one
also 9 times out of 10, people prefer the needle with heroin in it
> If those were real people making those comments, there is no hope for us.
they probably weren't.