Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT Images 2.0

openai.com

Post reply on HN
Print view View original

451–460 of 1001 posts

Re: ChatGPT Images 2.0

#451

Post by Planktonne »

NikolaNovak wrote: post #253
dakiol wrote: post #118

> On the flip side, there are hundreds of ways that these tools cause genuine harm, not just to individuals but to entire systems. Yeah, agree. I think it's the first time I'm asking myself: Ok, so this new cool tech, what is it good for? Like, in terms of art, it's discarded (art is about humans), in terms of assets: sure, but people is getting tired of AI-generated images (and even if we cannot tell if an image is…

While I agree with you, hacker news audience is not in the middle of the bell curve. I get this sounds elitist - but tremendous percentage of population is happily and eagerly engaging with fake religious images, funny AI videos, horrible AI memes, etc. Trying to mention that this video of puppy is completely AI generated results in vicious defense and mansplaining of why this video is totally real (I love it when vi…

HN is absolutely not more critical of AI output than the norm.

It's been true for various technologies that HN (and tech audiences in general) have a more nuanced view, but AI flips the script on that entirely. It's the tech world who are amazed by this, producing and being delighted by endless blogposts and 7-second concept trailers.

Re: ChatGPT Images 2.0

#454

Post by brynnbee »

rrr_oh_man wrote: post #375
neom wrote: post #252

Here is my regular "hard prompt" I use for testing image gen models: "A macro close-up photograph of an old watchmaker's hands carefully replacing a tiny gear inside a vintage pocket watch. The watch mechanism is partially submerged in a shallow dish of clear water, causing visible refraction and light caustics across the brass gears. A single drop of water is falling from a pair of steel tweezers, captured mid-splas…

Why would you consider this a good prompt?

My observations have been that image generation is especially challenged when asked to do things that are unusual. The fewer instances of something happening it has to train on, the worse it tends to be. Watch repair done in water fits that well - is there a single image on the internet of someone repairing a watch that is partially submerged in water? It also tends to be bad at reflections and consistency of two objects that should be the same.

Re: ChatGPT Images 2.0

#455

Post by 93po »

losvedir wrote: post #426
samiwami wrote: post #9

do they have anything similar to SynthID, or are they just pretending that problem doesn't exist? I know this is probably mega cherry-picked to look more impressive, but some of the images are terrifyingly realistic. They seem to have put a lot of effort into the lighting.

I feel like asking the image generators to mark AI images is the wrong way to go about it. It's like trying to maintain a blocklist. It seems better to me to have the major camera manufacturers or cell phones cryptographically sign their images as real.

I feel like this idea comes up often and in my opinion it doesn't solve anything. Take a picture of an AI image and you've made this approach useless. Which then goes to the argument of "well you'll see it's a picture of a picture" to which I will say there are plenty of ways to make this not appear so, and the ultimate form of this argument is that you can eventually project light directly into the photosensors, or otherwise hack the input between the photosensors and the rest of whatever digital magic that turns light into a JPG on your phone.

Re: ChatGPT Images 2.0

#457

Post by davecahill »

ritzaco wrote: post #28
tptacek wrote: post #24

Earlier quoted context omitted.

5.4 thinking says "Just right of center, immediately to the right of the HAM RADIO shack. Look on the dirt path there: the raccoon is the small gray figure partly hidden behind the woman in the red-and-yellow shirt, a little above the man in the green hat. Roughly 57% from the left, 48% from the top." (I don't think it's right).

I tried > please add a giant red arrow to a red circle around the raccoon holding a ham radio or add a cross through the entire image if one does not exist and got this. I'm not sure I know what a ham radio looks like though. https://i.ritzastatic.com/static/ffef1a8e639bc85b71b692c3ba1...

hilarious - i tried and got the same thing.

there was a very large bear in the first image; when asked to circle the raccoon it just turned the bear into a giant raccoon and circled it.

Re: ChatGPT Images 2.0

#458

Post by overgard »

kansface wrote: post #390
overgard wrote: post #216

Pretty mixed feelings on this. From the page at least, the images are very good. I'd find it hard to know that they're AI. Which I think is a problem. If we had a functioning congress, I wonder if we might end up with legislation that these things need to be watermarked or otherwise made identifiable as AI generated.. I also don't like that these things are trained on specific artist's styles without really crediting…

> If we had a functioning congress, I wonder if we might end up with legislation that these things need to be watermarked or otherwise made identifiable as AI generated.. Not a lawyer, but that reads as compelled speech to me. Materially misrepresenting an image would be libel, today, right?

Well, considering that AI generated content can't be copyrighted (afaik at least), I think we're in very different legal territory when it comes to AI creating things. While it's true that deepfakes could be considered libel.. good luck prosecuting that if you can't even figure out where the image came from.

The problem is it's all too easy to generate - you can't really do much about an individual piece of slop because there's so much of it. I think we need a way to filter this stuff, societally.

Re: ChatGPT Images 2.0

#459

Post by tfehring »

eclipticplane wrote: post #413
altcognito wrote: post #411

Earlier quoted context omitted.

.40 cents for high quality output is insanely cheap it is only going to get cheaper

> .40 cents Warning: Verizon math ahead.

In case anyone is unfamiliar with one of the most infuriating phone calls of all time: https://www.youtube.com/watch?v=MShv_74FNWU

Re: ChatGPT Images 2.0

#460

Post by anshumankmr »

dvt wrote: post #419
minimaxir wrote: post #234

So during my Nano Banana Pro experiments I wrote a very fun prompt that tests the ability for these image generation models to follow heuristics, but still requires domain knowledge and/or use of the search tool: Create a 8x8 contiguous grid of the Pokémon whose National Pokédex numbers correspond to the first 64 prime numbers. Include a black border between the subimages. You MUST obey ALL the FOLLOWING rules for th…

This is an amazing test and it's kinda' funny how terrible gpt-2-image is. I'd take "plagiarized" images (e.g. Google search & copy-paste) any day over how awful the OpenAI result is. Doesn't even seem like they have a sanity checker/post-processing "did I follow the instructions correctly?" step, because the digit-style constraint violation should be easily caught. It's also expensive as shit to just get an image th…

that is interesting cause I feel gpt-image-1 did have that feature.

(source: https://chatgpt.com/share/69e83569-b334-8320-9fbf-01404d18df...)

Post reply on HN