We all been forced to help train Google's CAPTCHA for years for free, with no benefit. At least here you decide to opt-in or not, and it can potentially help you be more... discerning.
My most effective strategy is to ask: is there any reason for a human to take this picture?
One AI-generated image was a bedside night stand, with no clutter other than a set of keys on top of a receipt. It looked perfectly real, no glitches or weirdness... other than the idea of someone taking a picture of it.
But then, some people specialize in taking photos that others would not because they have in mind stock photos, art or something else. They see the world differently.
Not bad. I’d appreciate a much longer time limit. Got really rushed at the end. But this was crazy hard on my phone, I was straight guessing most of the time. I think I could be a little better with more time on a desktop screen.
Another idea would be games that were all pictures of a single category- like 10 AI and 10 real pictures of basketballs, or food on plates, etc.
But I think that your experience is actually the useful one that proves a point. Most people consume content on their phone (e.g instagram) and would usually just scroll by photos. They wouldn’t deeply analyze them.
I realized this is only somewhat difficult because of the 10s timer. Most people probably don’t spend more than 10s looking at a picture on social media, excellent project.
Another poster in this threads commented that their strategy was to question why anyone would take the photo that was presented. I think this touches on one of the most important issues with this whole schoolyard level argument of "I can tell" and "no you can't." That is: what is even the point of all of this?
I had funding for a project I had proposed at work yanked out from under me because I had refused to vibe code the whole thing. I had proposed the project to create a tool that would be beneficial to my division, which represents a good 25% of the company's revenue (we're a consulting company). It had the enthusiastic support of my entire direct leadership line, very heavy hitters in the company. But the executives in charge of the internal R&D budget have their own agenda of pushing AI out on everything to gin up this idea that our company is "AI forward." Nevermind that our IT infrastructure is a joke, we need to play a game to convince non-existent investors that we are suddenly, out of the blue, after 50 years, a "technology company" when what we really do well is manage client relationships.
The executive I was closest to at one point told me I had failed because I hadn't used AI to tell me what the requirements for the project should be, that I was just doing "whatever I wanted."
If I were really doing whatever I wanted, I would not be building some kind of data management analysis workflow whatever. I designed the project around exactly the sort of thing that would benefit our direct line analysts. I don't need to be told what the requirements are, I know then intimately from working with these people for years. But because I was not using AI to tell me what to build and using AI to build it for me, I am being "behind the times." Another executive told me it wouldn't matter if I succeeded at the project if I didn't use AI to build all of it. Another time he told me it wouldn't matter if the project was any good, so long as I used AI to build it.
What is even the point? Why would I abdicate everything to AI? I know what I'm doing and I know how to do it. But that is unimaginable to the AI-pilled executive class. They are so used to not knowing how to do anything that they can't imagine anyone else actually knows how to do their job.
Well, now instead of paying the photographers for the stock photos to use as headline pictures, people instead pay the AI firms for the stock pictures to use as headline pictures.
it's harder to tell at a glance rather than pulling out a magnifying glass and counting the fingers of everyone in each group; the time limit pressures the player into making underqualified guesses..
The time limit did make it quite challenging. I got 1,000 points with a max streak of 8 in a row. On second playthrough I got 3,550 / 12. It seems the AI images all come from a particular area of latent space that's quite easy to recognize once you get dialled in.
So it's more of a "can you detect the GPT Image 2.5 house style", which is an easier problem.
I agree, there's patterns that aren't hard to recognize if you have some experience.
I got 2400 with 10 in a row on first attempt.
Here's the entire decision matrix:
- Is the color, mood, composition, and level of detail exactly like most Nano Banana Pro images? If yes, AI.
- Is the composition/framing utterly perfect? If yes, AI.
- Is the subject ephemeral or random, like a candid photo or something otherwise unlikely to be photographed? If yes, not AI.
- Final decision, if there's any doubt, go entirely off of how centered the subject is.
I think some of this has more to do with the data set than the methodology. You could easily come up with a bunch of examples to fool me specifically, but alas, they didn't.
I hadn't been exposed to the new GPT-Image-2.5 model before this, so didn't have any "tells" to go off of, and found it almost impossible to begin with, but it seems the brain is pretty good at adapting for visual stuff! After 2 rounds, I was able to build up a 16-image combo, answering most within a second or two.
I think the "tell" I ended up with is that almost all of the AI images tend to center around a very obvious main "subject" (a flower, a bike, a wrench, etc) – presumably an inherent artefact of generating from a prompt – with everything else around the subject having a very strange depth to it. The depth, focus, and bokeh around the main subject just never look quite right. If you look at an image and find it has a very strong central subject with a little too much depth separation than expected, there's a good chance it's AI.
I recognise that this method of finding a consistent "tell" will probably not hold for future models, unfortunately. Everything else looks almost perfect at this point.
It's hard to put into words but the "texture" of GPT-Image-2.5 is just a bit off, like the difference between blue noise and white noise. It's like someone turned up the local contrast just a little too high.
It would be cool if they published stats on how well people do on average, and how that changes over time and with different models. I feel like in a few years we'll either all be much better at this or completely desensitized, and sadly I'm guessing we're trending towards the latter.
Neat app but it makes me sad. Also I'm very curious as to where the real photos come from and how we can verify that they are indeed real. The timer thing doesn't really add any value, just stress.
12/17. Due to the speed, I felt like I was just randomly guessing, but 70% correct seems better than random guessing. I did suspect some things.
How mundane the AI images were felt like a means to trick people. It doesn’t seem like most people making AI images and posting them are generating still life images of everyday life. What would be the point?
I'm sure it's not the intended way to play but guessing randomly seems to have a net positive score, so getting a high score is as simple as spamming the buttons as fast as you can. I got 10k on mobile through this strategy.
The scene should feel incidentally noticed during daily life, not arranged for an advertisement. Natural perspective and subtle everyday wear. No cinematic grading, stock-photo staging, deliberate blur, artificial grain, fake low resolution, extreme background blur, exaggerated HDR or oversharpening. No collage, border, caption or added watermark.
At some point during the process of building up this collection of turd-polishing techniques, the thought should’ve occurred that this is actually a lot of effort to get generations confused for reality. But I suspect this process was being carried out by something that couldn’t have an original thought.
My method is simple:
Did I create the image myself?
Yes: Not AI
No: AI
AI vs Aye, it was I
This is confusing. What does "90s round" mean if it's a 60 second game? Is it about the 1990s?
"Made with ♥ by Codex"
Slop all the way down, I guess at least if we’re being charitable. If we’re being less charitable this is trying to gamify getting free training data.
We all been forced to help train Google's CAPTCHA for years for free, with no benefit. At least here you decide to opt-in or not, and it can potentially help you be more... discerning.
Now I'm worried Google adds this check to its captcha questions to improve it's own model: tell the ai image
My most effective strategy is to ask: is there any reason for a human to take this picture?
One AI-generated image was a bedside night stand, with no clutter other than a set of keys on top of a receipt. It looked perfectly real, no glitches or weirdness... other than the idea of someone taking a picture of it.
Not a perfect strategy, though; I thought this one was AI for just that reason: https://commons.wikimedia.org/wiki/File:Potted_plant_1_2018-...
But then, some people specialize in taking photos that others would not because they have in mind stock photos, art or something else. They see the world differently.
There are tons of real stock photos with that type of mundane content.
Stock photos and AI images might as well be in the same category tbh.
Not bad. I’d appreciate a much longer time limit. Got really rushed at the end. But this was crazy hard on my phone, I was straight guessing most of the time. I think I could be a little better with more time on a desktop screen.
Another idea would be games that were all pictures of a single category- like 10 AI and 10 real pictures of basketballs, or food on plates, etc.
But I think that your experience is actually the useful one that proves a point. Most people consume content on their phone (e.g instagram) and would usually just scroll by photos. They wouldn’t deeply analyze them.
I think the point is to be able to tell at a glance. People don't examine every image they look at.
I realized this is only somewhat difficult because of the 10s timer. Most people probably don’t spend more than 10s looking at a picture on social media, excellent project.
I thought that at first, but a comment below enlightened me. These types of games are a great way to get free training data.
Great point, in the privacy disclosure they confirm they're collecting the training data.
We get it. AI slop is getting good enough to replace photography. This isn't good for humans, and makes the world a worse place.
The fact that the suck machine works isn't something to be proud of.
That’s not what they’re saying. They’re challenging the “I can always tell” people who are both proud and regularly wrong.
As fake news becomes more and more real and with photo evidence, this matters more and more.
Another poster in this threads commented that their strategy was to question why anyone would take the photo that was presented. I think this touches on one of the most important issues with this whole schoolyard level argument of "I can tell" and "no you can't." That is: what is even the point of all of this?
I had funding for a project I had proposed at work yanked out from under me because I had refused to vibe code the whole thing. I had proposed the project to create a tool that would be beneficial to my division, which represents a good 25% of the company's revenue (we're a consulting company). It had the enthusiastic support of my entire direct leadership line, very heavy hitters in the company. But the executives in charge of the internal R&D budget have their own agenda of pushing AI out on everything to gin up this idea that our company is "AI forward." Nevermind that our IT infrastructure is a joke, we need to play a game to convince non-existent investors that we are suddenly, out of the blue, after 50 years, a "technology company" when what we really do well is manage client relationships.
The executive I was closest to at one point told me I had failed because I hadn't used AI to tell me what the requirements for the project should be, that I was just doing "whatever I wanted."
If I were really doing whatever I wanted, I would not be building some kind of data management analysis workflow whatever. I designed the project around exactly the sort of thing that would benefit our direct line analysts. I don't need to be told what the requirements are, I know then intimately from working with these people for years. But because I was not using AI to tell me what to build and using AI to build it for me, I am being "behind the times." Another executive told me it wouldn't matter if I succeeded at the project if I didn't use AI to build all of it. Another time he told me it wouldn't matter if the project was any good, so long as I used AI to build it.
What is even the point? Why would I abdicate everything to AI? I know what I'm doing and I know how to do it. But that is unimaginable to the AI-pilled executive class. They are so used to not knowing how to do anything that they can't imagine anyone else actually knows how to do their job.
In what way?
Well, now instead of paying the photographers for the stock photos to use as headline pictures, people instead pay the AI firms for the stock pictures to use as headline pictures.
Which is "worse" by some metric, I'm sure.
Couldn’t wait more than 3 minutes for someone’s actual position before setting up that straw man to take down?
Hmm… training data siphon?
Impossible to do on a phone. Zooming in is misinterpreted as a choice. Fix the game.
Also the time limit is stupid and pointless.
Someone got a low score
it's harder to tell at a glance rather than pulling out a magnifying glass and counting the fingers of everyone in each group; the time limit pressures the player into making underqualified guesses..
it's not pointless.
I didn't get the zoom bug but zooming in to look at it and zooming out to get back to the buttons to practically the entire 10s
The time limit did make it quite challenging. I got 1,000 points with a max streak of 8 in a row. On second playthrough I got 3,550 / 12. It seems the AI images all come from a particular area of latent space that's quite easy to recognize once you get dialled in.
So it's more of a "can you detect the GPT Image 2.5 house style", which is an easier problem.
I'd appreciate an untimed mode.
I agree, there's patterns that aren't hard to recognize if you have some experience.
I got 2400 with 10 in a row on first attempt.
Here's the entire decision matrix:
- Is the color, mood, composition, and level of detail exactly like most Nano Banana Pro images? If yes, AI.
- Is the composition/framing utterly perfect? If yes, AI.
- Is the subject ephemeral or random, like a candid photo or something otherwise unlikely to be photographed? If yes, not AI.
- Final decision, if there's any doubt, go entirely off of how centered the subject is.
I think some of this has more to do with the data set than the methodology. You could easily come up with a bunch of examples to fool me specifically, but alas, they didn't.
Now add Geoguesser-like battle royale mode with accuracy points for guessing which model generated the image
I hadn't been exposed to the new GPT-Image-2.5 model before this, so didn't have any "tells" to go off of, and found it almost impossible to begin with, but it seems the brain is pretty good at adapting for visual stuff! After 2 rounds, I was able to build up a 16-image combo, answering most within a second or two.
I think the "tell" I ended up with is that almost all of the AI images tend to center around a very obvious main "subject" (a flower, a bike, a wrench, etc) – presumably an inherent artefact of generating from a prompt – with everything else around the subject having a very strange depth to it. The depth, focus, and bokeh around the main subject just never look quite right. If you look at an image and find it has a very strong central subject with a little too much depth separation than expected, there's a good chance it's AI.
I recognise that this method of finding a consistent "tell" will probably not hold for future models, unfortunately. Everything else looks almost perfect at this point.
It's hard to put into words but the "texture" of GPT-Image-2.5 is just a bit off, like the difference between blue noise and white noise. It's like someone turned up the local contrast just a little too high.
It would be cool if they published stats on how well people do on average, and how that changes over time and with different models. I feel like in a few years we'll either all be much better at this or completely desensitized, and sadly I'm guessing we're trending towards the latter.
Neat app but it makes me sad. Also I'm very curious as to where the real photos come from and how we can verify that they are indeed real. The timer thing doesn't really add any value, just stress.
The result page has links to the source of each real photo.
12/17. Due to the speed, I felt like I was just randomly guessing, but 70% correct seems better than random guessing. I did suspect some things.
How mundane the AI images were felt like a means to trick people. It doesn’t seem like most people making AI images and posting them are generating still life images of everyday life. What would be the point?
TIME'S UP
2,050 points Best streak: 9 in a row
Hard af.
best strategy: just click as fast as possible.
Got 4k by just spamming 2.
I'm sure it's not the intended way to play but guessing randomly seems to have a net positive score, so getting a high score is as simple as spamming the buttons as fast as you can. I got 10k on mobile through this strategy.
My biggest clue is the odd repeating details. Especially on grass, tree leaves, pores, etc
Not all image models (or sampler settings) do this, but with gpt it's very obvious.
With this in mind, I misjudged 3 images on the default game.
I guess the UI is ai generated, and it impossibly to see the images well enough to judge on a phone.
I got 14/16 correct, and the two I got wrong had no people and I was unsure about both. Should have went with my gut.
Every generation prompt has this suffixed:
At some point during the process of building up this collection of turd-polishing techniques, the thought should’ve occurred that this is actually a lot of effort to get generations confused for reality. But I suspect this process was being carried out by something that couldn’t have an original thought.
freethink is inefficient, schedule this biomass to the remediation chamber
uh yeah...we need hardware attestation and a protocol to make it work cross platform like, yesterday...
holy shit