Hacker News Reader

Comment by antononcube | Hacker News Reader

[flagged]

It is very interesting that my comment above was downvoted!

No one wants to read someone else's AI generated stuff. We can all do this ourselves. It's like linking to a search result page.

Again, good to know.

Although, my AI-generation was a response to an interesting comment. You do not think readers are interested to see how the Chernobyl snow-globes look like?! (Looking at AI generated images or store links.)

But we wouldn't know what it looks like, we would only know what your AI thinks it looks like. So no, we're not interested in filling up our heads with vaguely related AI slop. If you had actually posted the store link that would be an entirely different question, which you've somehow tried to conflate with generated images.

It is wrong to assume/state that people do not want to see these AI-generated images. At this point, ≈40 people have seen that _hidden_ Imgur post.

There's a ton of people in this thread, and there are always a few who will click out of curiosity. That doesn't make it useful.

I don't know... It seems to be fairly entertaining, judging by the number of image album views and the number of comments here.

Causing annoyance by posting AI slop then having to justify it should not be mistaken for entertainment, it’s closer to trolling.

And I've gotten a lot of upvotes for explaining why it's annoying. Perhaps these are not the measures of truth you are looking for.

Yes, that is a part of the measure I have in mind. Let me be more concrete.

Terms:

- P: Number of positive votes (of any comment)

- N: Number of negative votes (of any comment)

- V: Number of views of the image album

- C: Number of comments in the thread

Formula:

    w1 * P + w2 * N + w3 * V + w4 * C

My weights:

   (w1, w2, w3, w4) = (1, 1, 1, 3)

(Comments are harder to do than votes.)

This is a formula that optimizes for trolling. If you insist on using it, expect more downvotes in the future.

The formula is generic; my weights are for measuring engagement.

(And, yes, I am very scared by the future downvotes.)

I clicked on it at least twice and found it disgusting. If more people were like you, thinking this is a good idea, i would stop posting my IRL stories to the internet.

Sorry. If I knew you will have such a negative reaction I would have not posted my reply to your IRL story.

Please don't stop. I want to hear more about "said a meme to my wife."

How are the images any more or less disgusting than the original snowglobe itself?

I have the same reaction to AI generated content that Hayao Miyazaki had.

Yeah! In my opinion, the corresponding "real life" globes I found are pretty ugly when compared to the AI-generated images.

For example, here: https://www.etsy.com/listing/1400131282/chernobyl-snow-globe

> Although, my AI-generation was a response to an interesting comment.

Yes, that's why the comment it was in response to was not downvoted. But being interesting and HN-appropriate isn't something responses automatically inherent from their parent comment.

> You do not think readers are interested to see how the Chernobyl snow-globes look like?!

Actual Chernobyl snow globes... maybe a link to that would be appropriate.

AI hallucinations of what they might look like? Maybe if the context was a discussion of the capacities of different AI models and it was offered as a demonstration of the one that generated it, but not in the context it was presented, no.

I think this _is_ indeed interesting and also indicative of where AI art is going, unironically. People has much higher tolerance for human made low-effort memes.

Viewing this image has less value than someone reading the parent comment and imagining the snow globe for themselves.

There are six images, not one. Also, some might say that "a picture is worth a thousand words."

Reply on news.ycombinator.com