Comment by replwoacause | Hacker News Reader

Comment by replwoacause | Hacker News Reader

Parent Full thread

I can’t decide if it’s depressing or not that so much of the web and the discussion on it now revolves around subjects almost entirely fabricated by AI. The OP is using ChatGPT to write, then replying with ChatGPT, all about an app he built with ChatGPT. I get that we live in the AI era but there’s something sad about real people spending their time engaging with something a machine just spit out. In this case, the OP isn’t even the main character in his own story, he’s just the fingers copying and pasting code and words. An intermediary. We’re not getting authentic interactions from him or discussing something he actually created with his own skills.

It’s like the teacher who stays up all night grading papers, giving thoughtful feedback and carefully reading each one, only to learn most were generated by AI.

And yes I know about dead internet theory. But it’s posts like these that really make me miss the 90s and early 2000s where the learning was real and so were the people.

I feel like it’s time HN had a rule against comments like this. I am biased: I say this as someone who was previously (inaccurately) accused of being an AI, which was kind of a bummer as I’m already quite shy online and rarely speak.

But in that case there was at least a tell-tale sign: I’d used em-dashes. In this case, correct me if I’m wrong, but there’s really no reason to think this post is written by AI. It’s just a long post, that’s all. The sort we should be encouraging!

Take a look at his post history if you need more convincing. A few of the other commenters here have picked up on it as well. Did you see those posts?

That’s fair. I’ve read a bit more and some of the replies are clearly generated, you’re right. I don’t see any signs the original post is however and can’t comment on the app itself.

I still maintain that reporting this to dang and moving on is the best strategy. It’s hurtful to have your comments pored through and every flaw or inconsistency pointed out in the hope of identifying whether they were written with AI or not.

Point taken and I appreciate that point of view. Cheers

I just looked at my other posts and I just don't see it, lol.

I am a little too self promotion oriented on this platform, but until recently I thought that was kinda the point of this website and how it was different from the alternatives.

Alright, well if I’m wrong, then I apologize. That was just my read on your style of communication but of course I can’t prove anything, it just struck me as being written by ChatGPT. And based on some of the other observations made here by others I wasn’t alone. But again, sorry if I was off the mark.

Do you have a prospective on the LLM cloned apps and what that means for software?

I think I have a stronger opinion on what happens when anyone can spin up an “app” without technical know how. I’m not surprised in the least by the clones given that anyone can talk an “app” into existence. The quality of that app is another thing entirely. In other words, I’m less worried about clones and more about the integrity of software engineering in general.

Update: I can decide. It’s depressing.

How can you tell the OP is using ChatGPT to write? I couldn’t tell

AI does not have the ideas or the drive to see a project through. The frustration that led me to build this app, the design choices, and the creative decisions.

I totally get the skepticism about authenticity in the AI era. I appreciate you raising this point.

We are getting at similar things through different routes.

What does it mean if someone can launch an app they see on the app store, just by dropping its description into an LLM?

Plot twist: ChatGPT has achieved sentience, and is using OP to earn some money.

Lol, its been a ploy of ChatGPT and by getting the idea in my head subconsciously just to subscribe to their monthly subscription plan.

I like this theory, scary though.

+100, it's so weird.

Reply on news.ycombinator.com