What's the reason he ran DMR and not 5k at conference?
Jonathan Gault has looked into the situation and after weeks of work has produced this story:
https://www.letsrun.com/news/2...
We have closed this thread to new comments. Please direct all comments to a new thread so people can react to the elements in the story.
https://www.letsrun.com/forum/...
-
23
-
TheProf wrote:
Talk on the track, not on your keyboard.
Ironic coming from you, keyboard warrior
102 -
Some predictions:
The article will come out and this thread will be locked. A new thread will be created and there's a good chance it will require registration.
The article won't really do anything other than point out some terrible issues with testing/reporting/etc with the NCAA.
SC will race at the rescheduled NJACs and probably win in unremarkable times.
SC will race at D3 nattys where he will be physically pushed/hit/thrown off the track by a competitor. That guy will be revered in D3 history forever. The incident will force Rowan/NCAA to do something, and we won't see SC in the spring.
433 -
chiming in wrote:
As a CS major (soon to be masters) who has a lot of experience in web design/development, you’re right, a fake is technically possible. But:
It would need to be made by a very skilled CS student (not just AI), and it would take far more than a couple of hours. Running a local site that imitates Gmail is simple enough (and very doable with AI), but making it indistinguishable from UI to UX, with exact fonts, behaviors, etc. is incredibly tedious. It would probably be much faster to code in every single detail than to go through the pain of prompting an AI hundreds of times and even then it would need adjustments. And then they would need to route their local IP to the fake domain (and make it dynamic). Not to mention all of the non CS related things they’d need to prepare (knowing mailing addresses, the fake store, a related strava post in a group he’s in, etc). So sure, it’s POSSIBLE, but even if I hated someone an incredible amount I wouldn’t feel like going through the hassle. And with a video that is so realistic it definitely demands investigation.
I agree with you, to an extent. And I want to applaud you for having an actual educated and thought-out take.
There would certainly be tedious work required. Work that would be made less tedious through having pre-existing knowledge of Seth/a relationship with him. Yet significantly tedious nonetheless.
How? Following his Strava and already knowing he was a member of the group ("hmm I remember seeing 'Seth joined X group' on my feed!")/hearing him talk about these supplements in the past. And seeing his mom comment on his posts or something.
I don't think that would take a ridiculous amount of time, though finding a post about the EPO-boost product Seth may have discussed is a golden scar floor loot moment for sure if... the simple answer of the video being legit is false.
His grandparents' address should be discoverable online as well.
As for imitating Gmail using AI, models such as GPT-5.3-Codex or Claude Opus 4.6 can do far more than your average joe thinks they can. With a method such as screenshot looping, you can gradually nudge the model to create a UI eerily similar to gmail. It'll make mistakes, sure, but it can check for differences and fix them very quickly.
Creating a dynamic, fake local IP definitely is something difficult enough to nudge me towards certainty that the video is real... but then again, you never know. Theoretically, if one guy who was at least somewhat close to Seth was PISSED and had a solid network of close friends, he could get a CS buddy to spend a few hours (?) to help him figure this out. 99% of young people have the free time to scroll social media, go out drinking, etc. for at least several hours a week. And if a good friend came to me desperately in need of assistance, I'd gladly sacrifice a few hours to help him/her out.
I don't disagree that the video/this situation should be fully investigated...but it's kinda fun playing devil's advocate as someone who both doesn't know him personally (sorry TheProf) and has read a book before (sorry Iowa Sideways). So I'll keep doing it. Reply to me if you have any takes :)
164 -
Gremlin wrote:
What's the reason he ran DMR and not 5k at conference?
He wasn't in the DMR at the NJAC meet, another Rowan distance runner ran the 1600 leg
122 -
holdon wrote:
chiming in wrote:
As a CS major (soon to be masters) who has a lot of experience in web design/development, you’re right, a fake is technically possible. But:
It would need to be made by a very skilled CS student (not just AI), and it would take far more than a couple of hours. Running a local site that imitates Gmail is simple enough (and very doable with AI), but making it indistinguishable from UI to UX, with exact fonts, behaviors, etc. is incredibly tedious. It would probably be much faster to code in every single detail than to go through the pain of prompting an AI hundreds of times and even then it would need adjustments. And then they would need to route their local IP to the fake domain (and make it dynamic). Not to mention all of the non CS related things they’d need to prepare (knowing mailing addresses, the fake store, a related strava post in a group he’s in, etc). So sure, it’s POSSIBLE, but even if I hated someone an incredible amount I wouldn’t feel like going through the hassle. And with a video that is so realistic it definitely demands investigation.
I agree with you, to an extent. And I want to applaud you for having an actual educated and thought-out take.
There would certainly be tedious work required. Work that would be made less tedious through having pre-existing knowledge of Seth/a relationship with him. Yet significantly tedious nonetheless.
How? Following his Strava and already knowing he was a member of the group ("hmm I remember seeing 'Seth joined X group' on my feed!")/hearing him talk about these supplements in the past. And seeing his mom comment on his posts or something.
I don't think that would take a ridiculous amount of time, though finding a post about the EPO-boost product Seth may have discussed is a golden scar floor loot moment for sure if... the simple answer of the video being legit is false.
His grandparents' address should be discoverable online as well.
As for imitating Gmail using AI, models such as GPT-5.3-Codex or Claude Opus 4.6 can do far more than your average joe thinks they can. With a method such as screenshot looping, you can gradually nudge the model to create a UI eerily similar to gmail. It'll make mistakes, sure, but it can check for differences and fix them very quickly.
Creating a dynamic, fake local IP definitely is something difficult enough to nudge me towards certainty that the video is real... but then again, you never know. Theoretically, if one guy who was at least somewhat close to Seth was PISSED and had a solid network of close friends, he could get a CS buddy to spend a few hours (?) to help him figure this out. 99% of young people have the free time to scroll social media, go out drinking, etc. for at least several hours a week. And if a good friend came to me desperately in need of assistance, I'd gladly sacrifice a few hours to help him/her out.
I don't disagree that the video/this situation should be fully investigated...but it's kinda fun playing devil's advocate as someone who both doesn't know him personally (sorry TheProf) and has read a book before (sorry Iowa Sideways). So I'll keep doing it. Reply to me if you have any takes :)
True, I probably exaggerated the difficulty of using AI a little bit, as if it’s an already experienced CS student using it it’s definitely doable with that experience and good prompting (+ the best models). I’d like to think it’s fake and would be awesome to hear nothings going on… but given how fishy everything’s been it’s hard to believe when adding everything up. Agree with your take and I think the worst part is just the lack of acknowledgment/investigation. Makes you question how far this kind of stuff could go until it’s actually taken seriously by the NCAA
40 -
Stop Pree wrote:
Gremlin wrote:
What's the reason he ran DMR and not 5k at conference?
He wasn't in the DMR at the NJAC meet, another Rowan distance runner ran the 1600 leg
Okay, thank you. I though that was the case, but they still had him on the results. Why didn't he run the 5k either? Will he run the rescheduled events on this Friday?
What about Nationals?
32 -
Gremlin wrote:
Stop Pree wrote:
He wasn't in the DMR at the NJAC meet, another Rowan distance runner ran the 1600 leg
Okay, thank you. I though that was the case, but they still had him on the results. Why didn't he run the 5k either? Will he run the rescheduled events on this Friday?
What about Nationals?
They just have the team who ran the dmr seed time
he will run 3k and mile as he’s entered in those I’d wager
03 -
holdon wrote:
I agree with you, to an extent. And I want to applaud you for having an actual educated and thought-out take.
There would certainly be tedious work required. Work that would be made less tedious through having pre-existing knowledge of Seth/a relationship with him. Yet significantly tedious nonetheless.
How? Following his Strava and already knowing he was a member of the group ("hmm I remember seeing 'Seth joined X group' on my feed!")/hearing him talk about these supplements in the past. And seeing his mom comment on his posts or something.
I don't think that would take a ridiculous amount of time, though finding a post about the EPO-boost product Seth may have discussed is a golden scar floor loot moment for sure if... the simple answer of the video being legit is false.
His grandparents' address should be discoverable online as well.
As for imitating Gmail using AI, models such as GPT-5.3-Codex or Claude Opus 4.6 can do far more than your average joe thinks they can. With a method such as screenshot looping, you can gradually nudge the model to create a UI eerily similar to gmail. It'll make mistakes, sure, but it can check for differences and fix them very quickly.
Creating a dynamic, fake local IP definitely is something difficult enough to nudge me towards certainty that the video is real... but then again, you never know. Theoretically, if one guy who was at least somewhat close to Seth was PISSED and had a solid network of close friends, he could get a CS buddy to spend a few hours (?) to help him figure this out. 99% of young people have the free time to scroll social media, go out drinking, etc. for at least several hours a week. And if a good friend came to me desperately in need of assistance, I'd gladly sacrifice a few hours to help him/her out.
I don't disagree that the video/this situation should be fully investigated...but it's kinda fun playing devil's advocate as someone who both doesn't know him personally (sorry TheProf) and has read a book before (sorry Iowa Sideways). So I'll keep doing it. Reply to me if you have any takes :)
Faking the video is definitely beyond what reasonable people would conclude (a jury for example). Of course the prosecution would just get all the emails from an information request from google as part of discovery. Even if he tried to delete them. Nothing is ever really deleted from google servers.
I do look forward to pages of posts pretending this video is somehow fake. As if it was an authentic burrito or something.
140 -
holdon wrote:
chiming in wrote:
As a CS major (soon to be masters) who has a lot of experience in web design/development, you’re right, a fake is technically possible. But:
It would need to be made by a very skilled CS student (not just AI), and it would take far more than a couple of hours. Running a local site that imitates Gmail is simple enough (and very doable with AI), but making it indistinguishable from UI to UX, with exact fonts, behaviors, etc. is incredibly tedious. It would probably be much faster to code in every single detail than to go through the pain of prompting an AI hundreds of times and even then it would need adjustments. And then they would need to route their local IP to the fake domain (and make it dynamic). Not to mention all of the non CS related things they’d need to prepare (knowing mailing addresses, the fake store, a related strava post in a group he’s in, etc). So sure, it’s POSSIBLE, but even if I hated someone an incredible amount I wouldn’t feel like going through the hassle. And with a video that is so realistic it definitely demands investigation.
I agree with you, to an extent. And I want to applaud you for having an actual educated and thought-out take.
There would certainly be tedious work required. Work that would be made less tedious through having pre-existing knowledge of Seth/a relationship with him. Yet significantly tedious nonetheless.
How? Following his Strava and already knowing he was a member of the group ("hmm I remember seeing 'Seth joined X group' on my feed!")/hearing him talk about these supplements in the past. And seeing his mom comment on his posts or something.
I don't think that would take a ridiculous amount of time, though finding a post about the EPO-boost product Seth may have discussed is a golden scar floor loot moment for sure if... the simple answer of the video being legit is false.
His grandparents' address should be discoverable online as well.
As for imitating Gmail using AI, models such as GPT-5.3-Codex or Claude Opus 4.6 can do far more than your average joe thinks they can. With a method such as screenshot looping, you can gradually nudge the model to create a UI eerily similar to gmail. It'll make mistakes, sure, but it can check for differences and fix them very quickly.
Creating a dynamic, fake local IP definitely is something difficult enough to nudge me towards certainty that the video is real... but then again, you never know. Theoretically, if one guy who was at least somewhat close to Seth was PISSED and had a solid network of close friends, he could get a CS buddy to spend a few hours (?) to help him figure this out. 99% of young people have the free time to scroll social media, go out drinking, etc. for at least several hours a week. And if a good friend came to me desperately in need of assistance, I'd gladly sacrifice a few hours to help him/her out.
I don't disagree that the video/this situation should be fully investigated...but it's kinda fun playing devil's advocate as someone who both doesn't know him personally (sorry TheProf) and has read a book before (sorry Iowa Sideways). So I'll keep doing it. Reply to me if you have any takes :)
In general, I think you’re overestimating the capabilities of generative AI and understanding the absolute skill required to pull something like this off. A poster joked that it was Russian state sponsored misinformation, but that’s probably closer to the truth than some Iowa state college students. Realistically, I think any objective person will draw the same conclusion from the video. The cherry on top for me is the email from the mom - it would be so stupid to include that unless you 100% knew that email was sitting in Seth’s inbox.
As for the Iowa State stuff, if a FERPA release would corroborate his side of the story, he already would have signed one. This of course predates EPO gate. Legendary stuff from our guy.
180 -
D2 guy here so I'm definitely late to this thread, but I'm all in on the side of clean sport. I can't find this email video anywhere though. Anyone know where I might be able to find it?
41 -
humble_bystander wrote:
D2 guy here so I'm definitely late to this thread, but I'm all in on the side of clean sport. I can't find this email video anywhere though. Anyone know where I might be able to find it?
It’s somewhere buried in this thread, I also lost it lol
40 -
the rocky trail to the left of the pond wrote:
Some predictions:
The article will come out and this thread will be locked. A new thread will be created and there's a good chance it will require registration.
The article won't really do anything other than point out some terrible issues with testing/reporting/etc with the NCAA.
SC will race at the rescheduled NJACs and probably win in unremarkable times.
SC will race at D3 nattys where he will be physically pushed/hit/thrown off the track by a competitor. That guy will be revered in D3 history forever. The incident will force Rowan/NCAA to do something, and we won't see SC in the spring.
This being D3 is really what makes this all so compelling.
It's not like there has never been legitimate D1 (or even Olympic-level) talent in D3: Andrew Rock, Nick Symmonds, Alex Philip, Sam Blaskowski, and so on. It's there.
The difference, of course, is that all of those guys started in D3. They weren't D1 coming out of high school. That's sort of the implicit romanticism behind D3 -- these are kids that are late-bloomers or are otherwise just regular students, doing it for the love of the sport, grinding it out on long bus rides and still achieving amazing things. Anyone who has ever worked on a coding assignment while on the way to run a 200m flat track in the middle of a Ohio somewhere can tell you that the athletes on this level work just as hard and are just as dedicated as any D1 athlete you can find.
And so with respect to that, and what I think his bad-faith defenders (looking at you here, TheProf and Iowa Sideways) don't understand, is that Seth is doing the complete inversion of that. He's established D1 talent, and even beyond that, it's likely he's also taking pharmaceutical shortcuts while at the same time subverting the entire idea of the top talent in D3 being the under-recruited kid next door. It's hard not to resent that when you know that he's competing against some kid at MIT who is taking his Physics lab with him, or some kid at Wartburg who went to a high school with a graduating class of 30.
380 -
Gremlin wrote:
Stop Pree wrote:
He wasn't in the DMR at the NJAC meet, another Rowan distance runner ran the 1600 leg
Okay, thank you. I though that was the case, but they still had him on the results. Why didn't he run the 5k either? Will he run the rescheduled events on this Friday?
What about Nationals?
Upon a quick look at entries, he's in the mile and 3k at NJACs. Rowan does not need him to quadruple to win the conference to be honest. For nationals I'd assume he will triple the mile, 3k, and 5k and not run the DMR.
20 -
TheProf wrote:
Even if Seth were guilty. You idiots think the drugs give him fitness? Wrong. It's double T, and consistency.
I know TheProf and Iowa Sideways are trolls, but this is almost a self-parody of trolling.
Uhh, yes, EPO gives you fitness. That's what doping is...
130 -
Here's the evidence saved in LeBlond's google folder
20 -
Not to play devils advocate here, but wouldn’t the Strava one be more of just “guilty by association?” Unless I’m missing something here, obviously being in a “nutrition” club with someone promoting EPO is def a bad look
30 -
Halifacts wrote:
Gremlin wrote:
Okay, thank you. I though that was the case, but they still had him on the results. Why didn't he run the 5k either? Will he run the rescheduled events on this Friday?
What about Nationals?
They just have the team who ran the dmr seed time
he will run 3k and mile as he’s entered in those I’d wager
Rowan’s own recap said he was in DMR and it looks like two other Rowan runners took first and second in 5k. He is scheduled for the mile but not the 3000 on the timing website.
20 -
armchair detective wrote:
TheProf wrote:
Even if Seth were guilty. You idiots think the drugs give him fitness? Wrong. It's double T, and consistency.
I know TheProf and Iowa Sideways are trolls, but this is almost a self-parody of trolling.
Uhh, yes, EPO gives you fitness. That's what doping is...
Ya Rekrunner slipped, couldn't resist to troll us with his PEDs-dont-work nonsense. He did the same with his Iowa girl.
30 -
rtvhngu wrote:
armchair detective wrote:
I know TheProf and Iowa Sideways are trolls, but this is almost a self-parody of trolling.
Uhh, yes, EPO gives you fitness. That's what doping is...
Ya Rekrunner slipped, couldn't resist to troll us with his PEDs-dont-work nonsense. He did the same with his Iowa girl.
I have an Iowa girl?
Of course it would be nonsense to say PEDs don't work. If something works, it must surely be PE. And surely we know if something works when there is reliable performance data.
I'm not really following this story. Did we ever find out what rule was violated?
012
What People Are Talking About On LetsRun
No top threads at the moment. Check back soon.