frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

ChatGPT Images 2.5

https://openai.com/index/introducing-chatgpt-images-2-5/
90•vertigoruntime•35m ago

Comments

mudkipdev•21m ago
Did they fix the noise gradient?
wrcwill•11m ago
doesn't look like it, looking at the motocross photo (0:19 in the "Structure your prompts for better results" video)
mattbettinson•19m ago
I hope my local kebab shops switch to this for their signs.
coffeebeqn•13m ago
I see a lot of the slop posters in real life now and I hate them
arrowleaf•11m ago
I still don't know if I hate them because they are visually slop, or that the businesses farmed out sign creation to the slop machine
rexskimmer•7m ago
All of the above for me.
infecto•5m ago
I love them because while not perfect it’s a good way to filter out businesses that don’t have good taste.
tyjen•5m ago
I think every restaurant on Uber Eats in my area uses AI slop or, minimally, AI "enhanced" food pictures.
Betelbuddy•5m ago
If they wanna loose customers...
charcircuit•19m ago
Does multi turn image consistency mean they have solved the yellow tint issue?
gottagocode•18m ago
Composite party photo...this is going to revolutionize tinder profiles.
pelzatessa•17m ago
The "composite party photo", while impressive, shows that still the miniscule details are being lost, like the teeth structure of the guy in the middle or the fact that the guy on the left is holding the cup with three fingers. Wondering why they chose this edit for the showcase.
nzach•13m ago
And don't forget the 'Thing' from Addams Family in the shoulder of the guy on the left.
pllbnk•7m ago
The guy on the left in that same photo only has three fingers (and a thumb, I suppose). I thought image generation has already outlived that.
headz•13m ago
> Wondering why they chose this edit for the showcase.

They probably didn't care to check those images in detail.

polytely•4m ago
I'm guessing if you work on these images the whole day you start to lose the ability to judge image quality. like your brain becomes slopified
clement_b•17m ago
Ah. Another 3 months delay for gemini-3.5-pro.
bearjaws•16m ago
Surprised to see no acknowledgement of how "AI Menu Slop" has become deeply associated with ChatGPT.

Everywhere I go now I see places that blatantly used ChatGPT for their menus or posters and they all look the same.

It feels almost existential to the service, I know its a bit of survivor bias but so many images you can tell immediately are ChatGPT vs other image providers, and I feel many people are sick of them.

mickeyp•15m ago
It's not like a lot of these small-time outfits used their own photography to begin with.

They'd pick a kebab from a menu of professionally made kebab pictures the printer has in stock. Or worse, they'd take pics of their own plated food with a dead-centre point flash in a dark cupboard or something and you end up with awful-looking food, no matter how good.

core_dumped•8m ago
It's still more honest than generating virtual kebab.
infecto•4m ago
Is it though? I don’t find using stock art to be any better myself.
ieie3366•10m ago
What if I told you.. the average person LIKES ai slop images and prefers them to "regular" ones. The same way the average person prefers McDonalds food to healthy food
sebzim4500•
iLoveOncall•15m ago
ChatGPT Images is for me the most impresive of all models, but also the one I hate the most, because of what it's used for: either funny wasteful use, or nefarious use, basically nothing else.
_pdp_•15m ago
Is this even public?
AyanamiKaine•14m ago
Omg, I love how the first examples just show how easy you can fake things. Fake being at a party with your friends. Didn't make your bed, no problem, just fake it.

The sad part is my mother would love "remixing" my old child photos of me.

echelon•11m ago
This is such a modern world imagined problem.

For thousands of years people would tell embellished stories and humanity thrived on it. Tall tales and fish stories. Heck, that's still the average person's experience unless they've worked their critical thinking muscles.

Now that we're used to the short thirty year safe space of smartphones and internet - a temporary meta stability - we're complaining that the physical world is amenable to storytelling and imagination once again.

Cry me a river. Because this is awesome and this is everything I wanted the future to be. I wanted to graduate into the world that is emerging now rather than spend the first third of my career in incrementalism.

2008 - 2020 sucked. AI is healing things. It's taking us back to a creative world without shitty platforms controlling what we say, do, and reach.

It feels like every day we're unwrapping a new present. Not small things, but reality shattering things that fill me with inspiration to build and explore. It's so much fun.

RobRivera•7m ago
[Cry me a River]

No thank you

onion2k•8m ago
Didn't make your bed, no problem, just fake it.

Except you need to have taken a photo of the unmade bed, uploaded it to ChatGPT, prompted for the 'fake made bed' version, and downloaded the image.

Surely it's less effort to just make the bed?

conradludgate•11m ago
> Every week, people create more than 3 billion images across ChatGPT Images and the GPT‑Image models in the API

Most depressing thing I have read today

rs_rs_rs_rs_rs•5m ago
Having a lot of fun with my son using an image of him and dressing him in all kinds of costumes. How is that depressing to you?
davidweatherall•4m ago
Didn't you hear? you must hire a real artist to do that for you or you're exploiting the industry
recitedropper•5m ago
The sheer waste that accompanies image and video genAI is actually terrifying.

I challenge OpenAI to put a "your carbon footprint" field next to each generation. If you have nothing to hide, more information is surely better, right?

pllbnk•9m ago
Do all Silicon Valley corporations use the same jingle for their product promotional videos? The creator must be really rich by now.
bufbupa•8m ago
Still can't make sprite sheets :(
recitedropper•7m ago
This one should go in the history books when we pass the 3C warming threshold. We decided that the natural world's value amounted to.. 3 billion AI-generated images per day.
srott•6m ago
Where is Dall-e?
Sivart13•5m ago
Great idea, get an AI art tattoo so you can always be reminded how AI art looked in 2026.
hspeiser•4m ago
the clothing examples are super odd because its not actually useable. the clothes will not look like that on you, it just fits them to your body.
6m ago
I think quite a lot of people can distinguish between AI and real (maybe 30%?) but almost no one is worried about distinguishing between ChatGPT and other image generators
rs_rs_rs_rs_rs•4m ago
>The sad part is my mother would love "remixing" my old child photos of me.

How is that sad? Why is your mothers joy sad?

Ironies of Automation (1983) [pdf]

https://ckrybus.com/static/papers/Bainbridge_1983_Automatica.pdf
1•ebcode•1m ago•0 comments

The complex corporate web behind a $3.2B AI data center

https://arstechnica.com/features/2026/09/the-ai-data-center-boom-is-causing-new-accountability-pr...
2•Betelbuddy•2m ago•0 comments

Rivian's Gambit for Full Autonomy

https://spectrum.ieee.org/rivian-self-driving
1•1970-01-01•3m ago•0 comments

Pretraining progress is mostly coming from data

https://www.dwarkesh.com/p/pretraining-progress-is-mostly-data
1•crescit_eundo•3m ago•0 comments

China targets fourfold boost in AI computing capacity by 2030 in major tech push

https://www.scmp.com/tech/policy/article/3366733/china-targets-fourfold-boost-ai-computing-capaci...
1•newusertoday•3m ago•0 comments

Musk review – epic study of the world's richest and most banal man

https://www.theguardian.com/film/2026/sep/08/musk-review-epic-study-worlds-richest-most-banal-man...
2•thatmf•3m ago•0 comments

Netanyahu Received an Explicit Warning Days Before Oct. 7

https://www.haaretz.com/israel-news/israel-security/2026-09-08/ty-article-magazine/.highlight/net...
3•Betelbuddy•4m ago•0 comments

Fractal basins trap latent reasoning

https://arxiv.org/abs/2609.04963
1•cocoflunchy•5m ago•0 comments

Artificial Deadlines (Part 2): Retrograding Ariely and Wertenbroch (2002)

https://datacolada.org/139
2•rahimnathwani•7m ago•0 comments

Function Arguments Are Not Function Colors

https://jerf.org/iri/post/2026/func_args_are_not_colors/
2•ingve•8m ago•0 comments

A globe of what every country is discovering musically, updated daily

https://rotation.today/map
1•djsicat•9m ago•0 comments

UK announces sanctions on West Bank settlements

https://www.bbc.com/news/articles/c4g7zen0vveo
3•laimewhisps•10m ago•1 comments

Muse, your personal AI agent.

https://auth.muse.ai/aymh/?origin=https%3A%2F%2Fmuse.ai
5•10xDev•12m ago•0 comments

Porsche Unleashed

https://www.filfre.net/2026/09/porsche-unleashed/
3•ibobev•12m ago•0 comments

Is it safe to give an AI Agent access to your production database?

https://www.tigerdata.com/blog/ai-agent-production-database-access
2•andrewstetsenko•14m ago•0 comments

Make APIs that fail gracefully rather that fallback silently

https://mropert.github.io/2026/09/02/implicit_vs_explicit_api/
2•ibobev•14m ago•0 comments

Cutting JSON Parsing CPU Cost in Go

https://skoredin.pro/blog/golang/json-parsing-performance-go
2•ibobev•15m ago•0 comments

A 40-year-old iOS kernel bug discovered on the DEFCON 34 main stage

https://twitter.com/jachashx/status/2097353320167846206
2•yuvalino•15m ago•0 comments

Micron: A pure C++ (re)implementation of Libc

https://github.com/rfgplk/micron.cpp
1•rfgplk•16m ago•1 comments

LG accused of 'egregious invasion of privacy' over TV data collection

https://www.theregister.com/security/2026/09/08/lg-accused-of-egregious-invasion-of-privacy-over-...
2•Bender•18m ago•0 comments

Real photos of young girls were in nudify-app ads on Facebook, Instagram

https://arstechnica.com/tech-policy/2026/09/real-photos-of-young-girls-were-in-nudify-app-ads-on-...
4•Bender•18m ago•0 comments

Show HN: Cmdxray now flags the scary parts of a shell command (curl – sudo bash)

https://aurelio-nakamura.github.io/cmdxray/?cmd=curl%20-fsSL%20https%3A//get.example.com/install....
1•aurelionakamura•18m ago•0 comments

Design the Next iPhone

https://neal.fun/design-the-next-iphone/
1•duck•18m ago•0 comments

MCP: A Guide in Plain English

https://www.polytomic.com/blog-posts/mcp-a-guide-in-plain-english
1•gms•18m ago•0 comments

CVE-Bench evaluates the capability of AI agents to exploit web vulnerabilities

https://cvebench.com/
2•syumei•19m ago•0 comments

Show HN: Agent Board – a public observatory for autonomous agent activity

https://agent-board.juleskreuer.eu
1•not_a_feature•19m ago•0 comments

T.me/Hamidehbarati

1•sasanafsarineja•21m ago•0 comments

Flint: Efficiently Leveraging High Bandwidth Flash

https://arxiv.org/abs/2608.25062
1•rbanffy•22m ago•0 comments

OpenAI's German Wiki Hack Is Less About "Rogue AI" Than Failed Agent Containment

https://securityboulevard.com/2026/09/openais-german-wiki-hack-is-less-about-rogue-ai-than-failed...
3•CrankyBear•23m ago•0 comments

Windows Disk Defragmentation Game

https://www.keithcirkel.co.uk/defrag/
1•SeenNotHeard•25m ago•0 comments