Apply an ai like this with more detections to a typical instagram „models“ stream and you‘ll see that almost all pictures are altered. This could maybe be a good wakeup call for the young generation which is seeing the bodies as problematic because they are not as perfect as other people on instagram
It is usually more insidious than that. Almost everybody know that Instagram picture are heavily altered, it doesn't prevent us from building an image in our head of what we "should" look like.
It would be a great idea to legally force social networks to automatically apply this kind of AI and show the results in the feed for the reasons you mentioned.
Clear strawman fallacy here. I did not say 'forbid'. I said that it should be made clear that the images are doctored.
Besides, there is a big difference between make-up, that anyone could theoretically apply to look like someone in real-life and what people use photoshop for in the images we are talking about.
I would call it reductio ad absurdum. But that's not the point as I wasn't refuting any arguments you made (you didn't any), but making fun of what I see as a knee-jerk reaction of calling for a new law on every minor occasion.
Btw, your final para: do you mean that only privileged people can apply "beatification" effects on their phitos? FWIW make-up is much more expensive.
After a week - at most - none of the millions of users sees the "photoshopped image" anymore, it'll become ubiquitous. Mostly due to the fact that each and every image will have that label / text
So while obviously edits are a huge issue with body image. It's very important to understand that when you look at almost any IG picture, whether it is altered or not, you're most likely looking at the best possible aspects of the person that they specifically chose to show.
You're only seeing the best parts of someone while comparing them to your worst parts. And when every single picture is specifically crafted to only show the best parts, you'll get a very unrealistic idea on what the person actually really looks like.
One day I was sitting on the bus. There was someone in front of me crafting an Instagram post, then somebody else in front of them just chilling listening to music not paying attention to anyone or anything. I watched while the person in front of me took a few different selfies from a couple angles, added some filters then sit there crafting a tale about dude in front who was doing absolutely nothing, don't even think he noticed anyone else at all on the bus, and how he'd said something to her and her selfie was the reaction. It took a bit of time too. She thought about it, edited it and changed it a bunch. It was some effort. All just crafted to get reaction.
It was entirely fabricated, not a single thing was true about the post. Then I watched while she eagerly refreshed checked ever 15 seconds for likes and sympathetic comments.
It was kind of funny to watch, but sad at the same time. I can't understand such a strong need for validation that you'd do such a thing.
Now the team in charge of "Face-Aware Liquify" has to improve their algorithm so that the edits won't be recognizable.
That would mean better warping ("more real looking"), which is overall a great thing, even if it just ups the ante for the for the "Edit detector" team.
Basically, the red team has to hack into the image instead of a program - definitely a good way for Photoshop to evolve.
(Though I realize that this detection would be useful for much more than just the liquify tools.)
>Trust in what we see is increasingly important in a world where image editing has become ubiquitous – fake content is a serious and increasingly pressing issue.
I keep on reading such statements, but still don't feel like this is a real (increasing) problem. Pictures could be faked is a believable way 100 years ago already, and I cannot recall a single instance where new, fancy AI-driven fakes were any more of an issue than classic photoshopping. In my opinion, it's all just FUD.
100 years ago you would have needed an expert, specialized equipment and several hours. 10 years ago, you still needed an expert and probably an hour at least. Soon, there may be tools anyone can use, feed in some photos of you and a description of what you want to see. It will be easy to generate fake profile pictures that look like you, for instance. Your voice will be easily cloned on the phone as well, and eventually, even video chats could be fake.
Scammers could find out when you're on a trip from social media, reach out to your family, ask for emergency money because of a stolen wallet, etc. Still some effort for the scammers, but eventually, you might be able to automate the whole scam with a bot. The bot doesn't need to be that smart, it just needs to follow a script with a handful of variants, like a preplanned dialogue tree.
This is something that could be done before as well by just impersonating someone. The voice just needs to sound believably similar, which is not too hard with a sufficiently bad connection. And yet such attacks don't seem common at all.
Sure, you can impersonate someone (or even anyone). A botnet can impersonate everyone, and make robocalls to everyone.
If you're doing a targeted attack, you might need a significant chance of success - if you can target 10 000 users, you might get away with a 1% success rate, and still make a significant amount (say 60 times 1 000 USD in "emergency" funds).
I think the issue is scale. Keeping track of one doctored image with a missing Yezhov next to Stalin is easy.
Keeping track of myriads and millions of doctored images is harder. Even if you have a hash database or thumbprint database to keep track of all of them, someone is going to think they’re miscategorized or attribute the classification to politics or soft censorship. etc.
A common way that bots are detected on twitter is that they re-use other users' avatars. This is pretty quickly being replaced by https://thispersondoesnotexist.com/ (and others).
25 comments
[ 1.9 ms ] story [ 63.0 ms ] thread(and yes, somebody already turned this into an app)
This here?
https://www.reddit.com/r/iOSProgramming/comments/e08sps/woke...
Costs 3,49 Euro in my local App store.
Besides, there is a big difference between make-up, that anyone could theoretically apply to look like someone in real-life and what people use photoshop for in the images we are talking about.
Btw, your final para: do you mean that only privileged people can apply "beatification" effects on their phitos? FWIW make-up is much more expensive.
It brings to mind the cloning bans based on Hollywood logic where they assume clones have all memories and the same exact scars and birthmarks.
Here's a good example https://www.reddit.com/r/Instagramreality/comments/ciw4a3/do...
Or just look at someone like Chessie King https://www.instagram.com/chessiekingg/
You don't need to edit photos to give an unrealistic image of how your body is. You just need to know some basics of posing, angles and lighting.
The same applies to men as well
https://www.reddit.com/r/Instagramreality/comments/djvv1b/th...
So while obviously edits are a huge issue with body image. It's very important to understand that when you look at almost any IG picture, whether it is altered or not, you're most likely looking at the best possible aspects of the person that they specifically chose to show.
You're only seeing the best parts of someone while comparing them to your worst parts. And when every single picture is specifically crafted to only show the best parts, you'll get a very unrealistic idea on what the person actually really looks like.
It was entirely fabricated, not a single thing was true about the post. Then I watched while she eagerly refreshed checked ever 15 seconds for likes and sympathetic comments.
It was kind of funny to watch, but sad at the same time. I can't understand such a strong need for validation that you'd do such a thing.
That would mean better warping ("more real looking"), which is overall a great thing, even if it just ups the ante for the for the "Edit detector" team.
Basically, the red team has to hack into the image instead of a program - definitely a good way for Photoshop to evolve.
(Though I realize that this detection would be useful for much more than just the liquify tools.)
I keep on reading such statements, but still don't feel like this is a real (increasing) problem. Pictures could be faked is a believable way 100 years ago already, and I cannot recall a single instance where new, fancy AI-driven fakes were any more of an issue than classic photoshopping. In my opinion, it's all just FUD.
Scammers could find out when you're on a trip from social media, reach out to your family, ask for emergency money because of a stolen wallet, etc. Still some effort for the scammers, but eventually, you might be able to automate the whole scam with a bot. The bot doesn't need to be that smart, it just needs to follow a script with a handful of variants, like a preplanned dialogue tree.
If you're doing a targeted attack, you might need a significant chance of success - if you can target 10 000 users, you might get away with a 1% success rate, and still make a significant amount (say 60 times 1 000 USD in "emergency" funds).
Keeping track of myriads and millions of doctored images is harder. Even if you have a hash database or thumbprint database to keep track of all of them, someone is going to think they’re miscategorized or attribute the classification to politics or soft censorship. etc.