Forum
>
General / Nonfiction
>
Thread
Log in |
Register
her recent rise in numbers could be probably due to her editing her entire body with AI. A clear example is this vid she posted today
it looks like you're right about that specific vid, but having seen her on live streams etc. i don't think this is a serial issue that warrants a complete blacklisting from the main page. face filters are pretty common on douyin, xhs, and chinese social media generally so unless the issue is especially egregious and shows up in every post she makes online, i don't think it warrants a total blacklisting either. she's got one of the best physiques and overall looks on the site, so this would be a pretty big L imo if she remains blacklisted. ofc though, i get why it's an issue -- but having seen her posts across multiple platforms for a pretty long time, i really don't think it's as chronic or as problematic an issue as with Guo Hui for example.🥲
I didn't soft blacklist her because of the face editing, but rather the body editing.
understood -- i just had never noticed it as being such a chronic issue with her posts. aside from the one example linked here i can't imagine she's been editing her physique in every single post or even the majority.
understood -- i just had never noticed it as being such a chronic issue with her posts. aside from the one example linked here i can't imagine she's been editing her physique in every single post or even the majority.
Welp, after reading thru this thread's mentions of YH it seems it has indeed been a common issue.
The soft-blacklisting's disappointing but it's very likely for the better.🥲
Yan Hong is too gorgeous to fully ban. Oh such a shame she does this, literally Eun Hee Kang all over again.
Yeah my thought process is this:
She posts a video where there's an obvious AI glitch, where she or her editor failed to cover up an AI edit.
Even with said AI edit, she isn't obviously bigger than in previous videos.
We know she already uses a filter for her face.
All this combined makes it vanishingly unlikely that this is the first AI edit she's ever done. She's likely been doing it for a long time. This is just the first time she messed up in an obvious way.
(Every time this happens, it's a huge shame because the woman doing it already has an amazing physique; she just feels the need to cheat to go even further than that for some reason)
Yeah that's what's brutal about it. She's already been facing down haters on douyin etc. for being too jacked. So it's kinda sad to know that she's been faking her physique for validation in the opposite direction. Same unfortunate dynamic just a different flavor.
Society.😔
Honestly I think this is getting a little out of hand. I get that some people don't want any edits at all, but like, there are so many of us that just don't care. Can we have a global toggle in our cookies or user settings or something that allows these soft-blacklisted models to appear in the main site?
This thread has been an exercise in subjectivity overruling objectivity.
understood -- i just had never noticed it as being such a chronic issue with her posts. aside from the one example linked here i can't imagine she's been editing her physique in every single post or even the majority.
Spotting fakes isn't for the casual observer, but that doesn't mean it isn't there. The distinction isn't if you're casually noticing, but if you're applying that casual observation as a proactive denial when someone that scrutinized it more closely has flagged something suspicious. Sure, that works in the opposite direction where someone casually flags something as suspicious when it isn't, but this isn't a topic for casual opinion in either direction.
Yan Hong is too gorgeous to fully ban. Oh such a shame she does this, literally Eun Hee Kang all over again.
The Soft Blacklist isn't a full ban. They're still on the site, you can fav them, search them, afaik you can still add uploads and receive notifications. The Soft Blacklist just means they won't be on the main page. No need for catastrophizing.
Honestly I think this is getting a little out of hand. I get that some people don't want any edits at all, but like, there are so many of us that just don't care. Can we have a global toggle in our cookies or user settings or something that allows these soft-blacklisted models to appear in the main site?
Why ask Chainer to code a new feature that you can easily accomplish by starting a thread which lists everyone in the Soft Blacklist? If you want a site with a main page of cartoonish reality distortions, I recommend DeviantArt.
This thread has been an exercise in subjectivity overruling objectivity.
Spotting fakes isn't for the casual observer, but that doesn't mean it isn't there. The distinction isn't if you're casually noticing, but if you're applying that casual observation as a proactive denial when someone that scrutinized it more closely has flagged something suspicious. Sure, that works in the opposite direction where someone casually flags something as suspicious when it isn't, but this isn't a topic for casual opinion in either direction.
everyone's entitled to an opinion. either way, it's a huge L. she really didn't need to go the AI route to prove herself.🥲
I think it's important people don't get the pitchforks out and start a witch hunt. There is a kinder way to go about it
but this. also this.💯
This thread has been an exercise in subjectivity overruling objectivity.
I haven't seen anyone refuting the evidence.
As I understand it we found one instance where there is a glitch which shows physique enhancement. (And face filters which are applied constantly, but that can be spotted by the casual observer often enough).
Based on this some might think it's something which is maybe done now and then, but the scale and frequency might be limited.
Others might think it's a Lance Armstrong kind of operation where a team of professionals are busy day and night creating and manipulating (enhancing) the footage, and to detect and cover up glitches that may remain. And that this was one rare slip-up, which uncovers the entire operation.
But going by the evidence we don't know this. We don't know which tools they have used, we don't know the scale and frequency of the practice. Given what we have found we have to deal with it in a certain manner.
My opinion (yes, an opinion) is that Chainers line of thinking here is reasonable, and the soft-blacklisting is appropriate, even though I am a bit disappointed. And she doesn't even need it, to edit her stuff. She has a great physique. She won a pro-competition (of which we have the photo's directly from the source) and qualified for the Olympia. Anyways, this last part is me coping again 😅
I haven't seen anyone refuting the evidence.
As I understand it we found one instance where there is a glitch which shows physique enhancement. (And face filters which are applied constantly, but that can be spotted by the casual observer often enough).
Based on this some might think it's something which is maybe done now and then, but the scale and frequency might be limited.
Others might think it's a Lance Armstrong kind of operation where a team of professionals are busy day and night creating and manipulating (enhancing) the footage, and to detect and cover up glitches that may remain. And that this was one rare slip-up, which uncovers the entire operation.
But going by the evidence we don't know this. We don't know which tools they have used, we don't know the scale and frequency of the practice. Given what we have found we have to deal with it in a certain manner.
My opinion (yes, an opinion) is that Chainers line of thinking here is reasonable, and the soft-blacklisting is appropriate, even though I am a bit disappointed. And she doesn't even need it, to edit her stuff. She has a great physique. She won a pro-competition (of which we have the photo's directly from the source) and qualified for the Olympia. Anyways, this last part is me coping again 😅
That middle part read like an AI wrote it. 🤔
Not sure how you interpreted my "subjectivity" comment as "refuting evidence" given that I expound on the theme with individual replies.
Subjectivity as in wanting to turn the GWM Galleries into an AI and "morphing" free-for-all because 3 users are indifferent and object to any efforts of the site to curate for quality and for real physiques.
Subjectivity as in cognitive dissonance by not seeing or believing that Yan self "morphs" or claiming it's an isolated anecdote despite her entire GWM gallery matching the exaggerated muscle mass of the concrete evidence. AKA, we DO know the scale and frequency of the practice, and we know it's AI tools even if we don't know the exact software. That's like saying it's not really photoshopped because we don't know if it was literally Adobe Photoshop or another program of similar purpose. It's enough to know it was manipulated and the exact tool is irrelevant.
It doesn't matter that she doesn't need it. It only matters that she manipulates her own media because that's the only relevant criteria.
And the all purpose "everyone gets an opinion" angle in response to a very narrowly directed comment regarding concrete evidence isn't the gotcha either of you think it is.
No. Evidence based criteria isn't open to opinions.
I’ve scrubbed through this video some and watched it to look for what you’re talking about, and I really can’t see what you’re talking about. Can someone be more specific about what they’re seeing that indicates an ai filter over this? Please tell me where in the frame to look and what time to look at it.
I DO see some weird shadowing on her shoulder as she turns her body at the start, that just looks like typical LED lighting flicker. But I don’t see an overlay suddenly blip into existence from one frame to the next.
If anyone can be very specific about what they’re seeing see, I’d appreciate it. I feel like I’m usually better than most people at catching that sort of stuff and I just don’t see it.
I’ve scrubbed through this video some and watched it to look for what you’re talking about, and I really can’t see what you’re talking about. Can someone be more specific about what they’re seeing that indicates an ai filter over this? Please tell me where in the frame to look and what time to look at it.
I DO see some weird shadowing on her shoulder as she turns her body at the start, that just looks like typical LED lighting flicker. But I don’t see an overlay suddenly blip into existence from one frame to the next.
If anyone can be very specific about what they’re seeing see, I’d appreciate it. I feel like I’m usually better than most people at catching that sort of stuff and I just don’t see it.
All details you requested are in my post here. I made a gif animation (mp4 for GWM) of 3 screenshots to make it as obvious as possible. Scrub through that because it drops the first (necessary for context) screenshot when it loops.
I’ve scrubbed through this video some and watched it to look for what you’re talking about, and I really can’t see what you’re talking about. Can someone be more specific about what they’re seeing that indicates an ai filter over this? Please tell me where in the frame to look and what time to look at it.
I DO see some weird shadowing on her shoulder as she turns her body at the start, that just looks like typical LED lighting flicker. But I don’t see an overlay suddenly blip into existence from one frame to the next.
If anyone can be very specific about what they’re seeing see, I’d appreciate it. I feel like I’m usually better than most people at catching that sort of stuff and I just don’t see it.
Her left arm from 00-01 the bicep area compresses and then flickers back to normal size.
That middle part read like an AI wrote it. 🤔
Thanks. But I don't need AI for bad writing 😉
Subjectivity as in wanting to turn the GWM Galleries into an AI and "morphing" free-for-all because 3 users are indifferent and object to any efforts of the site to curate for quality and for real physiques
I understand that you don't want this, and for that matter, I don't want that either. And I also want to add that I appreciate your efforts in this quality control and trying to detect the edits/glitches. But users are free to have their own opinion about this matter. In the end it's Chainer who decides.
Subjectivity as in cognitive dissonance by not seeing or believing that Yan self "morphs" or claiming it's an isolated anecdote despite her entire GWM gallery matching the exaggerated muscle mass of the concrete evidence. AKA, we DO know the scale and frequency of the practice, and we know it's AI tools even if we don't know the exact software. That's like saying it's not really photoshopped because we don't know if it was literally Adobe Photoshop or another program of similar purpose. It's enough to know it was manipulated and the exact tool is irrelevant.
This is your logical deduction based on the evidence, and I tend to agree for a large part, that is: I also don't think it's an isolated event, but happens more often. You seem to be very sure about the scale and frequency, I am not going to debate you on this, because I think it is likely too, but that's something different than evidence. The evidence is the glitch you found.
It doesn't matter that she doesn't need it. It only matters that she manipulates her own media because that's the only relevant criteria.
I said I was disappointed and that she has a great physique regardless and therefore doesn't need to do this. This is my personal feeling about this. I never said this is a reason we should be less strict to the relevant criteria. (That is: we agree on this)
No. Evidence based criteria isn't open to opinions.
I don't know how you come to that conclusion that I am debating that. Based on the evidence we can have opinions how we deal with it, though. And as I pointed out, I think the soft-blacklisting is appropriate.
Thanks. But I don't need AI for bad writing 😉
I understand that you don't want this, and for that matter, I don't want that either. And I also want to add that I appreciate your efforts in this quality control and trying to detect the edits/glitches. But users are free to have their own opinion about this matter. In the end it's Chainer who decides.
This is your logical deduction based on the evidence, and I tend to agree for a large part, that is: I also don't think it's an isolated event, but happens more often. You seem to be very sure about the scale and frequency, I am not going to debate you on this, because I think it is likely too, but that's something different than evidence. The evidence is the glitch you found.
I said I was disappointed and that she has a great physique regardless and therefore doesn't need to do this. This is my personal feeling about this. I never said this is a reason we should be less strict to the relevant criteria. (That is: we agree on this)
I don't know how you come to that conclusion that I am debating that. Based on the evidence we can have opinions how we deal with it, though. And as I pointed out, I think the soft-blacklisting is appropriate.
Yes. You're correct. My last reply... the majority of my replies are logic driven. Logic provides clarity and distinction.
I'm sure about the scale and frequency because that's been irrefutably determined by the evidence. The evidence is MORE than the glitch. You're treating the glitch as evidence of an isolated example of AI enhancement only. It's also evidence of Yan's baseline muscle mass that can easily be compared to all her GWM gallery media. It's true that I'm eyeballing it. It's also true that my eye is trained by 30+ years as an enhancement artist. If you disbelieve, that's acceptable and the onus is on me to back my claims up. My claim is the baseline muscle mass established by the evidence is many times smaller than Yan's body of work, so I'd randomly sample stills of that upper arm in a similar position and flexion, remove the backgrounds so only the body was visible, color shift the control image (the unenhanced baseline) to blue, and I'd color shift the random samples to red. I'd scale and rotate to overlay the samples over the control, run an "unblur" filter to strengthen the visibility of the contours, lower the opacity of both images to roughly 15%-33% range depending on brightness and contrast between the layers, and I'd lay both over a third completely white background layer. The net effect of those steps should achieve a four color high contrast comparison image with white background, red sample, blue control, and purple overlap. If no enlargement exists, it will all be purple on a white background, while enlargement will appear as the red smoking gun concrete evidence visual aid that I can eyeball without the elaborate steps above as my aid. To quote Black Widow: "I've got red in my ledger, I'd like to wipe it out."
We're both disappointed that she has a great physique regardless: "I hate needing to report this because her body (and arms) are amazing..." - phenoms, May 16, 2026
I wasn't putting words in your mouth, I was drawing a cold, objective, logical distinction between relevant and irrelevant.
Yes, you can have any opinion you want about how Chainer wants to deal with issues, but Chainer has final say regardless.
There are times when opinion is valuable in trying to improve things, but at other times it's loud background noise that can drown out objective reality, facts, and evidence. When evaluating for fakes, that's an evidentiary process with no room for the noise of opinions. The question then becomes if the opinion is a contribution to improve the process, or noise drowning out an objective evaluation of evidence.
Thanks for the further clarification, and it seems you've built a solid case. Keep up the good work.
Yan hong has been clearly editing her face for a couple of years now but her recent rise in numbers could be probably due to her editing her entire body with AI. A clear example is this vid she posted today, you can see the filter flickering at the start of the vid on her arm… https://www.girlswithmuscle.com/2832496/