The fact that discord is actually apologizing to some people tells me that they did have a human reviewer that got lazy and decided to use AI to do their job
Someone has to confirm AI flagging that content. Either that mod did a mistake, or they hated Marvel Rivals and thought they'll play a little prank on the fans @user-to7ds6sc3p
One time I called the mods on a predator in the official Minecraft discord server. About 4 months later it was flagged for child safety. Calling the mods on a predator, was flagged as predatory behavior. My account was suspended for 1 day and it would have this strike on record for 2 years, so if Discord falsely flagged the account again it'd just get banned. Ever since then I haven't trusted Discord at all to actually enforce their rules properly. I stopped being a nitro subscriber that day.
Doing a good thing, makes your account flagged, therefore you should just ignore the predators so your account doesn't get flagged due to the "guilty by association" addage and just be a spectator. ie MALICOUS COMPLIANCE.
I think I know why. If a server gets taken down for any reason, it'll punish everyone that was ever in it. Even if you were the one who made the report, you're still counted among the people who joined the server.
@ In this case, the server wasn't at fault. They dealt with the person promptly and thanked me. Discord just saw it later and falsely flagged it, then gave me a copypaste email when I complained.
Had literally the same thing happen to me. Stumbled upon a weird server by an underage guy trying to make a dating server. I called it out, told him to delete the server, because it was getting abused by the predators very heavily. He promptly banned me and apparently a couple of days later Discord force deleted the server and I got flagged for child safety. And, well, before that I got a very mysterious flag with no explanation whatsoever, so who knows. Both happened a couple months between each other around May last year. Discord moderation is as terrible as always
@@Jesin00 Yes, but you'd think programmers would do an absurdity check on that. Like if whatever happens is based on a date, and that date is before the creation of the Internet, don't do that thing
@@williamdrum9899 You do realize that it defaults to January 1st 1970 for a reason, right? It's caused by an incorrect integer, which makes the epoch time become 0 due to a conflict between the actual epoch time. 0 is Janurary 1st 1970. This problem is so insignificant and difficult to correct that it's useless trying to correct it.
@@ParrotPentesteryes but at the same time just compare the Unix time codes. if the timecode is long before Discord, or skype, or Apple or Micosoft hell the only "computer" you could get would be a minicomputer (as in, not the size of a shed) with a teletypewriter (or a graphical desktop with a Xerox or Bell machine, or a terminal for everything else) so yeah, if it predates any meaningful timecode, and is, itself, closer to 0 than to the current timecode, then it should be null and void like a ban from 2009 would make sense because its code is closer to ours than to 1970 but any form of glitch with timecodes like that should not happen on the modern web whether its 1/1/1970 or sometime in 1992; a good programmer would have a simple check to say "this isn't ours, our timecodes only exist above 2012" I haven't programmed in well over a decade and it's trivial for me to do that. like the easiest thing to do in programming if [current time]
Account Recovery for Discord, Facebook, Epic, etc is already so bad that people are losing their 16 years old active account and can't appeal to get it back, and now over that you get falsely banned. Nice.
Yep, lost an account that had 2fa on it because they want me to use an authenticator app and the codes change so fast you can't type them in if you're using a screen reader. It's insane!
@@SarafinaSummersHave you tried using Ente Auth? It lets you see the next code that is going to be generated. So that's double the time you have to input a code.
@@SarafinaSummers If you're in the US, this is actually something you could contact an attorney about making a complaint under the Americans with Disabilities Act.
Tbh I’d take this over complaining and making discord literally force their employees to look at CP every time they get a report related to it. You wouldn’t like doing it so it’s not right to force others to do it either, the best alternative is a machine… though even just pixelating the images/videos just enough to make it less graphic would be better for reviewing them.
@rarehyperion someone being falsely flagged as a child predator is definitely not the better solution, it legitimately ruins lives to be falsely accused of such a vile act, also they could actually account for the bots having false positives by actually having a appeal system where a human can review what the ai flagged themselves if the person request a appeal.
Makes me remember the story of the artist who got his Google account nuked. Not warned, not banned, nuked entirely like it never existed. Gmail, Drive, UA-cam - everything GONE. Said artist took extreme precautions to avoid getting their speedpaints age restricted, deleted or demonetized, yet lost their channel overnight. Last time I checked the strongest theory was that someone could've dropped a picture to a shared Drive folder that Google's internal filters flagged as CSAM. Edit: there was other case, when a father got his Google account nuked for sending pictures of his kid to a doctor for a legit reason (can’t remember if it was a dog bite or an accident, the details escape me). He sued Google, but I don’t know how it ended.
Saito Naoki. He was a big channel that did art tutorials and speed paints. He had a drive where people could submit their work for critique and advice, which is where people believe the ban came from.
@@ajddavid452Fuck Google Edit: This is the first time I'm hearing that CSAM and any other pics of a minor in that way is different. I always thought it was no excuses but maybe not.
fun fact according to corporations censorship is ok and so is ok breaking hands of freedom of all sapient beings.[my comment was deleted] if they can do that what you describe, then they arent amongst the flourishing nor withering away; they are amongst the withered ones.
Happened to my friend. Literally sent an image of a peach (the actual fruit) and he then got permbanned. Later that day he told me that he wont talk much because the police is investigating in his house. Discord flagged it as cp btw
this is genuinely ridiculous. it makes the app so so so unsafe. if it can be proven that AI is being used they need to be sued into the ground. it states directly on their website that a person reviews it. i say this because as you mentioned, being banned for CSAM used to immediately tell everyone you were a predator. now its meaningless. and they already lost a ton of money recently.
The worst part about this is not the fact that there was one high profile case of false positives on their detection system... But the fact that that likely means there was a lot more false positives and people getting banned because of it and that nobody knows about because it didn't effect that many people and so it didn't get a lot of attention.
Plus, A person who has good content doesn't say that their content is "better" - in that case you'd be being arrogant. Plus, you are a gachatuber, what should have I expected xd
YES!!!!! I GOT FALSELY BANNED THE SAME. DAY AS THIS! My thought process was that it was a hacker from a scam I encountered because I got banned for a "financial scam," but this just brings it to a WHOLE new level.
Same. I can't imagine a worst job than having to moderate content on discord or telegram or Facebook. The amount of things these people watch in a day of work must be absolutely traumatizing and nightmare-fuel. Unless they like it, which I don't think makes the situation better.
There's groups like "Exodus Cry" that try to categorize all vaguely mature content out there as "sex-trafficking" so I wouldn't put it past one of those types to have weaselled their way into discord's moderation team to label everything even remotely sexy as being the most egregious thing they can. Flooding the database would (in their perspective) have the added bonus of just removing ALL mature content from the internet, legal or otherwise.
Is there any proof of this? Sorry not being rude I just want to know if they're really like that before I label in my mind, what seems to be a good cause, as bad
unjust and ignoble worlds logic: 'if you do this, you get ban' just and noble worlds logic: 'if you do this and it can be reversed and removed, you have a [timeframe] to remove it, othervise face a ban in the face'
5:49 Imagine the mod that was forced to watch that just f*cking hates Marvel Rivals and decided to flag it as CP just out of spite LMAO. Wouldn't even put it past them tbh.
Craziest thing about the appeals not working is that a week ago I got someone banned for the same reason, and they appealed it and got their account back.
I won't be surprised if there will be an entire AI based discord server that got self reported by itself (AI), and then got banned the very next day with this reason: *AI GENERATED CONTENT*
I think it's because in the bottom right of the video it says "teen," so this AI saw inappropriate video and then saw the word "teen" and assumed it was a teenager in the video. That's just my guess :,)
@ Hello Dakota, I never once stated I was covering for corporations. I was actually saying their AI is very dumb and associated the age rating for the video game to be related to the contents of the video, as that's my guess. That is clearly stupid because it's very recognizable that the AI was wrong, hence the whole point of this video. The AI is and was not capable of checking accompanying text because it is dumb and a cheap option for moderation, which is why this happened in the first place. The AI made a mistake because the corporation messed up on checking it's faults.
As a human moderator and community manager of many years, I can definitely say that computers won’t be doing our job for years. It just doesn’t do it right and AI will even admit when I have conversations with it that it doesn’t understand nuance. It sees boobs and thinks oh no, not safe for work… but the context of that is that the woman was wearing a dress and it was on the floor of the Emmy‘s… Yeah, not NSFW rip AI rip
AI can only do so much and it is stupidly quick how much diminishing returns you get with AI-based systems. This is because the technology is iterative since it enforces itself with more data but then it suffers the same issues as generative AI where it begins to "hallucinate" and flag patterns so generally that anything even remotely similar would trigger a report. AI currently is pretty good at recognizing something very specific but anything more general it started to fail because it lacks any higher level processing capability (as in, actual intelligence). All companies will very soon realize that limitations to AI systems will make them lose productivity so they will be begging people to come work for them again.
@@TheOz91 Begging them to work for moderation? They would probably need to hire hundreds of thousands of people to do so, and even then they would need high pay because they have to look at CSAM and inappropriate content AI moderation is not getting replaced with how large these kinds of platforms have become, however, there might be soon a breakthrough on machine learning for specifically detecting inappropriate content, that is only if these companies invest tons of resources towards figuring out how to do that, and they obviously won't unless the law pressures them
@@silly.spongus Sure but the tech is still going to be inferior compared to people. For that reason, they will still need to have human "supervisors" because despite the progress of ML and AI in general, a lot of false positives will still exist. Again, AI will never be truly intelligent. I do support the use of AI in these sort of jobs but just like every job that companies are trying to replace people with, the diminishing returns will force them to hire back at least a few skilled staff members.
@@silly.spongus They make enough money selling your data and from idiots giving them money from subscriptions. The least they could do is pay *real people* money to make their shit less dystopian.
@@gairisiuilthe AI and human probably thought the character was a minor based on the "T for teen" thing in the bottom right most likely no lmfao. yet somehow real crap sneaks through and doesn't get flagged
They were already using AI over a year ago, I got hit with a ban from their AI over a false match over a year ago & it took like 2 months to finally get a real person to review it & get my account back. It took way too long to get an actual person to look into it, I've not resubscribed to nitro since, I still use discord but I have no trust in it anymore. When a AI can instantly perma ban you for something like that & it takes months of spamming support to get a real reply that is a big issue.
5:09 A human reviewed my appeal on the copyright strike on my video about a song from Mojang's creator safe playlist in a Minecraft video. Within 5 seconds they had received my appeal, reviewed the ~2minute clip and answered, as a supposed "human".
Is it possible that instead of Discord picking up this post, flagging it, and sending it to the NCMES, that maybe another company did this and when it hit the NCMES database Discord AI acted on it? The clip of the website shown in this video shows that "once items hit the database all other applications can see it" (paraphrased, from memory)
Not _impossible_ , but it likely would have come up in other forums as well... Marvel is, I'm pretty sure, not a terribly small thing, and if X(Previously Twitter) _isn't_ using AI moderation... We'd likely have seen some fallout from that as well.
ai will not be fully replacing moderation anytime soon edit:if you see ANY bots report them for spam edit 2:is ntts or youtube eating the bot replies idk but its good either way
I know someone how got suspended because he joined a server which had a secret channel that he didn't have access to and someone posted a guide to make a bomb in that channel. Discord basically banned everyone on that server not just the on that sent it
As someone who does moderation myself, AI should never be in charge of actually placing bans / infractions. The most they should ever do is flag to a queue for humans to look over.
That doesn't necessarily mean 12 minutes with one woman, it could be 1 minute for 12 different women, or even 1 minute but 12 different times with 1 woman
Situations like these remind me to not put all my eggs in one basket and not to put my communication methods in the hands of one company. This goes for other tech things as well, use multiple operating systems or at least know how to and keep a copy just in case something happens to one.
I’m sorry Rain world… here of all places 0:32 I NEVER see anything about Rain World when it’s not just the game but I see it in the “Discord ai is banning people for no reason video” ah yes makes perfect sense
@Letaryu cleverbot esp, nerdcubed did a video on it way back and one of the things it said was "*gives master water*" and that was uncomfortable then and it's uncomfortable now, considering how cleverbot works
That gif where the guy eats popcorn was ACTUALLY from a CSAM video, so that mass ban made way more sense. Sue Storm in Marvel Rivals is canonically in her 60s and used technology to make herself look young. She's literally the oldest character in the game. 💀 It's not CSAM. 💀
So what I'm hearing is, if someone inserts a popular gif or other non-offending image into illegal materials and that gets pickup by AI, hilarity ensues
I actually had an old child safety ban overturned after going to court over it. What happened was that a server I was in got raided, and the raiders posted and reported the stuff they posted. It was so much that it couldn't be cleaned up in time and the entire server got nuked. Discord, in court, tried to argue that I was responsible for it as I was in the server when it happened. Judge ruled in my favor, and we settled for damages and an unban. Discords actions have always been questionable, but it's the first time it got stuff wrong on such a scale. Hope they will fix the system and especially allow appeals in the future, because the default no appeal possible is just not good enough.
I got banned one year ago for being admin on a banned serwer. It was banned because of child safety with no further explanation. This was a discord on making desserts. Mostly people posting pictures of cookies they made and discussing what the best brownie recipe. I have no freaking idea why it got banned.
I hate it when this happens, now a new excuse for creeps try to get away with it. Fortunately, Discord did send out the emails with free nitro gifts as an apology for the mistake, which means Discord did take up responsibility. Discord sending out those emails caused a huge copypasta about Discord's email system got compromised or a massive phishing attack impersonating Discord.
I got sent actual CP in my dms yesterday and then I reported it, guy didn’t get banned and now people are getting banned for forwarding “CP” even though it’s just a normal image.
Same. A dude that showed me his litteral files of pics and videos. (He was edgy, telling me he liked to study the anatomy. Like bruh) He also didn't get banned. I can't tell what it takes to get a user banned for actually committing a cime instead of being thirsty for a Marvel Rivals skin. It's truly pathetic.
A friend of mine was caught in the Germany bans and it took him over a month to get his account back and only got it back, after I talked to a friend who knows a Discord employee...
The whole NCMEC getting flooded with this stuff is why people reporting lolicon to the NCMEC despite them literally telling people to stop. Animations are not missing or endangered children. Don't do that. Remember that every time you report lolicon to them, you are actively responsible for a child being hurt.
Dude, my partner had their discord account permanently suspended for this... we were freaking out, had no idea what was wrong, years of memories lost lol.
create own system of operating, own sites and own nation and history and declare yourself as above all human bc you have a righterous morality superior than other people's minds
@@kaydog890 don't know if you're just stupid or genuinely don't know because you don't touch grass yourself, but people can use discord as a messaging application AND have a life of going outside and touching grass at the same time. maybe don't be fucking stupid next time you comment, hope you get better
@@fujiwaranonekobiodrando1257 Most likely because the actual term can't be said in a video because of monetary reasons? I'm not entirely sure but it's why a lot of channels have to tread lightly because they know UA-cam is sensitive about certain topics but like I said I don't entirely know for sure
i talk to the same 5 people every day, i didn't even open the marvel rivals server even tho i am in it, didn't share it and still got banned. like... come on...
Ironically I had the same thing happen (not for the Rivals thing) but as part of the AI sweeping a bunch of accounts. I recommend messaging them on X (both Discord and Discord Support) because there’s at least one actual person who checks it. I’m saying this if you’ve already gone through the other means. I got my account reinstated, and the issues fixed. Just took them like a week and fights with the Clyde bot. But yeah, I wouldn’t be surprised if it’s just AI at this point. Their system is so broken.
in roblox never say a word and you will be safe in discord never post anything and you will be safe and i thought microsoft will ruin discord no they ruin themselves
now i hate AI above everything cause companies use it as the worst excuse possible and con artist use it as their golden ticket but AI in video game is completely different and rain world AI is one of the best i seen that actually until i get on raindeer and that thing is the reason why the grass stops a run
man had to bring up breaking child saftey, i some how broke it once and i dont even know why, then he has to say its the worst thing to be flagged for :((.... anyways great video
I actually got banned for the Marvel Rivals incident forwarding the message with my comment of “the snowbunny propaganda grows”. Thankfully Discord reverted it shortly and gave me three months of nitro for compensation Legitimately thought it was because of a free subscription I attached to a prepaid card that had like a dollar on it. Nice to know it wasn’t that IG
AI makes sense for mass reports, however AI should only handle the obvious, but there should be a team of people to double check reports that the AI might be unsure of.
I have no idea about this case specifically, and my ban was not involved in it. However, I find it amusing that this banning spree happened THE SAME DAY I got falsely banned out of nowhere. I was just minding my business, and then I got forcibly logged out and said my account was permanently suspended for "financial scam," whatever tf that implies. I got my account back, but it was such a headache to deal with because my thought process is that a hacker who scammed me months ago got into my account at *some point* between that scam and now and got reported. Thankfully, discord responded to me after countless bot responses (unlike they did at the time I was asking for help from that scam) but it should have never happened in the first place.
I'm on the verge of writing a script to mass-delete every message I send about a month after I sent it, just to prevent some random AI from looking at it and deciding that I should cease to exist.
@@richardmillhousenixon the default mod log servers have doesn't actually save messages people deleted, only who deleted them and when (seriously that's incredibly stupid for a multitude of reasons, mod logs shouldn't save deleted messages). The mod logs you see are all done using third party bots that just repost what people send after it's deleted. Still, discord does probably keep your messages for a time even after you deleted them (it's a waste of space to keep them forever, Discord is already bleeding enough money as is), so all deleting really does is save you from someone reporting your old messages.
NTTS, moderation AI predates generative AI and I'm pretty sure Discord and other platforms has been using AI moderation for at least awhile now, platform like UA-cam probably a decade by now. Machine learning and algorithms have been used for many things for many decades. Unless you know they've changed that up their moderation system significantly, we shouldn't assume beyond what we can already surmise. This new prob could easily be a bug, but also humans have made errors like this too. I feel as if you're unfairly blaming AI, AI that you don't even know exists within Discord beyond the industry standard algorithmic automated moderation filtering. It's the same old cycle, we bIame the new unknown and spooky tech for problems of yesterday. We just keep shifting who's fauIt it is.
discord not banning actual CSAM and banning people for THIS is so dumb. this does NOT need AI, there's ways to check image hashes automatically and from what I can recall, there's services that provide those hashes to filter out CSAM. AI should be a last resort with how unreliable it is, and even then, as that one user in the video said, there should be a good way to treat the false positives.
A better approach is use AI to guess whether it is offending, THEN send it to a human moderator, they'll determine if they get banned and then they can even train the AI to be better.
If I'm not wrong, many big social media platforms switched to using AI to detect child pornography as it is less controversial than a human looking at what potentially could be child pornography.
That "free nitro gift links" email being real is absolutely wild -- people are absolutely going to send phishing emails that look exactly like that one now.
I would much rather pay money to a service that has ACTUAL HUMANS running moderation of said service than a service that uses AI that EASILY gets abused by some of the worst people on the planet. Facebook, UA-cam, Twitter, Discord, every social media platform, every single website or application that uses AI as a major source of moderation all deserve to lose a shit ton of money for that reason alone. If I have to pay more for humans to make a good service, EVEN IF SOME THINGS TAKE LONGER, I'll fucking do it. When I pay money for a service, I expect PEOPLE to be moderating and handling the service. NOT an AI backend.
>bans for child porn
>it's neither child nor porn
good to see ai is as good as always.
love it
It's replacing human jobs already, too.
We are a doomed society
@@Diddy_biggestfan09 🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖🤖
To be fair, the character model is kinda sexualized... like everything is nowadays...
But yeah, nothing banworthy
@@heidtb6746its not a sexualized model its literally toned down lol
The funniest part is that the character they banned us the oldest woman in the game
I mean, Sue Storm is literally a married woman
Technically black widow is 200 in rivals
weirdly we consider fictional characters' age??
@@fujiwaranonekobiodrando1257???
@@fujiwaranonekobiodrando1257well yea. thats like saying watching a "animated child” isn't pedophiliac
The fact that discord is actually apologizing to some people tells me that they did have a human reviewer that got lazy and decided to use AI to do their job
I would be really worried if people that are reviewing potential cp or csam have enough freedom to let ai look at that stuff.
Someone has to confirm AI flagging that content. Either that mod did a mistake, or they hated Marvel Rivals and thought they'll play a little prank on the fans @user-to7ds6sc3p
NoTextToSpeech literally said they replaced the humans with AI to save money, therefore, what human reviewer checked it?
@@heavyweaponsguy6284 and its well known that he works at discord and knows everything thats going on there
Discord's email was hacked. I don't play Marvel Rivals, and they sent me a phishing email that looks identical.
Discord fails to ban actual cp but bans this
Edit: its possible that a lawsuit could be made just the same as roblox
Discord is ran by furries who consume said cp all the time btw
real
@@starisgone it's ran by furries so
@@slon-z1v what does that have to do with anything?
@@slon-z1v One of their former staff was a furry and predator. It was back when cub content was allowed on the platform
One time I called the mods on a predator in the official Minecraft discord server. About 4 months later it was flagged for child safety. Calling the mods on a predator, was flagged as predatory behavior. My account was suspended for 1 day and it would have this strike on record for 2 years, so if Discord falsely flagged the account again it'd just get banned.
Ever since then I haven't trusted Discord at all to actually enforce their rules properly. I stopped being a nitro subscriber that day.
Doing a good thing, makes your account flagged, therefore you should just ignore the predators so your account doesn't get flagged due to the "guilty by association" addage and just be a spectator. ie MALICOUS COMPLIANCE.
They refuse to take a server down for federal offences, so no surprises there.
I think I know why. If a server gets taken down for any reason, it'll punish everyone that was ever in it. Even if you were the one who made the report, you're still counted among the people who joined the server.
@ In this case, the server wasn't at fault. They dealt with the person promptly and thanked me. Discord just saw it later and falsely flagged it, then gave me a copypaste email when I complained.
Had literally the same thing happen to me. Stumbled upon a weird server by an underage guy trying to make a dating server. I called it out, told him to delete the server, because it was getting abused by the predators very heavily. He promptly banned me and apparently a couple of days later Discord force deleted the server and I got flagged for child safety.
And, well, before that I got a very mysterious flag with no explanation whatsoever, so who knows. Both happened a couple months between each other around May last year.
Discord moderation is as terrible as always
got banned for something I did 55 years ago. discord didn't even exist 55 years ago, me neither.
Jan 1, 1970 is the default date on many computer systems when you forget to put in a date.
@@Jesin00 Yes, but you'd think programmers would do an absurdity check on that. Like if whatever happens is based on a date, and that date is before the creation of the Internet, don't do that thing
@@williamdrum9899 You do realize that it defaults to January 1st 1970 for a reason, right? It's caused by an incorrect integer, which makes the epoch time become 0 due to a conflict between the actual epoch time. 0 is Janurary 1st 1970. This problem is so insignificant and difficult to correct that it's useless trying to correct it.
@@ParrotPentesteryes but at the same time just compare the Unix time codes. if the timecode is long before Discord, or skype, or Apple or Micosoft
hell the only "computer" you could get would be a minicomputer (as in, not the size of a shed) with a teletypewriter (or a graphical desktop with a Xerox or Bell machine, or a terminal for everything else)
so yeah, if it predates any meaningful timecode, and is, itself, closer to 0 than to the current timecode, then it should be null and void
like a ban from 2009 would make sense because its code is closer to ours than to 1970
but any form of glitch with timecodes like that should not happen on the modern web
whether its 1/1/1970 or sometime in 1992; a good programmer would have a simple check to say "this isn't ours, our timecodes only exist above 2012"
I haven't programmed in well over a decade and it's trivial for me to do that. like the easiest thing to do in programming
if [current time]
LMAO
Account Recovery for Discord, Facebook, Epic, etc is already so bad that people are losing their 16 years old active account and can't appeal to get it back, and now over that you get falsely banned. Nice.
i lost my 9 y/o account during the first wave :)
Yep, lost an account that had 2fa on it because they want me to use an authenticator app and the codes change so fast you can't type them in if you're using a screen reader. It's insane!
@@SarafinaSummersHave you tried using Ente Auth? It lets you see the next code that is going to be generated. So that's double the time you have to input a code.
@ Never heard that one before. RiP
@@SarafinaSummers If you're in the US, this is actually something you could contact an attorney about making a complaint under the Americans with Disabilities Act.
TLDR: Discord doesn't care, they trust machines more than you
Ur not cool for this btw. My contents better... 😈
Vrchat gooner
this
Tbh I’d take this over complaining and making discord literally force their employees to look at CP every time they get a report related to it. You wouldn’t like doing it so it’s not right to force others to do it either, the best alternative is a machine… though even just pixelating the images/videos just enough to make it less graphic would be better for reviewing them.
@rarehyperion someone being falsely flagged as a child predator is definitely not the better solution, it legitimately ruins lives to be falsely accused of such a vile act, also they could actually account for the bots having false positives by actually having a appeal system where a human can review what the ai flagged themselves if the person request a appeal.
Makes me remember the story of the artist who got his Google account nuked. Not warned, not banned, nuked entirely like it never existed. Gmail, Drive, UA-cam - everything GONE. Said artist took extreme precautions to avoid getting their speedpaints age restricted, deleted or demonetized, yet lost their channel overnight. Last time I checked the strongest theory was that someone could've dropped a picture to a shared Drive folder that Google's internal filters flagged as CSAM.
Edit: there was other case, when a father got his Google account nuked for sending pictures of his kid to a doctor for a legit reason (can’t remember if it was a dog bite or an accident, the details escape me). He sued Google, but I don’t know how it ended.
Who was it?
Saito Naoki. He was a big channel that did art tutorials and speed paints. He had a drive where people could submit their work for critique and advice, which is where people believe the ban came from.
The father was taking pictures of his son’s genitals to send to his pediatrician because he had a rash. Pictures that the doctor asked for.
google even reported the father to the police and the police concluded that no this isn't csam and Google STILL didn't give him his account back
@@ajddavid452Fuck Google
Edit: This is the first time I'm hearing that CSAM and any other pics of a minor in that way is different.
I always thought it was no excuses but maybe not.
fun fact they can also report you to the police for this false ban
but most of the time they probably dont do that, so it is a paper tiger
but seriously, why ppl nowadays are so sensitive? ois it due to communism??
thats scary and fucked up
fun fact according to corporations censorship is ok and so is ok breaking hands of freedom of all sapient beings.[my comment was deleted]
if they can do that what you describe, then they arent amongst the flourishing nor withering away; they are amongst the withered ones.
ok that's just fucking illegal lmao
Happened to my friend. Literally sent an image of a peach (the actual fruit) and he then got permbanned. Later that day he told me that he wont talk much because the police is investigating in his house. Discord flagged it as cp btw
1:57 maybe the video gets flagged as child endangerment because its written "TEEN" on the screen
That's probably why the ai fucked up!
That's hilarious ngl
imagine thinking a video is cp because it says rated T for teen
@ ai doesn't think, it sees teen and its got a 🍑, now it will think it is cp.
its a dogwhistle
Aritifical lncompetence
real😭🙏
Genuine Stupidity
lmao
Born from TikTok island
@rarehyperion TikTokers invented LLM, that's why it's so stupid and buggy! lol, just kiddin: they ain't intelligent enough to do AI research.
this is genuinely ridiculous. it makes the app so so so unsafe. if it can be proven that AI is being used they need to be sued into the ground. it states directly on their website that a person reviews it. i say this because as you mentioned, being banned for CSAM used to immediately tell everyone you were a predator. now its meaningless. and they already lost a ton of money recently.
They can be sued regardless, they made false accusations against thousands of people. That's slander, not to mention false accusations are illegal.
The worst part about this is not the fact that there was one high profile case of false positives on their detection system... But the fact that that likely means there was a lot more false positives and people getting banned because of it and that nobody knows about because it didn't effect that many people and so it didn't get a lot of attention.
I wonder who asked u... My content is way better 😺
Who is the individual above my comment? Hey you! You are like the clouds, when you are gone It's a good day :3
Plus, A person who has good content doesn't say that their content is "better" - in that case you'd be being arrogant. Plus, you are a gachatuber, what should have I expected xd
Gotta love the bot replies lmao
YES!!!!! I GOT FALSELY BANNED THE SAME. DAY AS THIS! My thought process was that it was a hacker from a scam I encountered because I got banned for a "financial scam," but this just brings it to a WHOLE new level.
Discord AI banned me by sending a friend image, thinking it was "CP". Their support only answer with a bot, and I can't get my account back
xD
Same here
What did you send
@@bogeymanbearprobably an irl image
@bogeymanbear a bro pic without a shirt
God, I can't imagine being the poor soul who has to moderate that content on discord...
Same. I can't imagine a worst job than having to moderate content on discord or telegram or Facebook. The amount of things these people watch in a day of work must be absolutely traumatizing and nightmare-fuel.
Unless they like it, which I don't think makes the situation better.
There's groups like "Exodus Cry" that try to categorize all vaguely mature content out there as "sex-trafficking" so I wouldn't put it past one of those types to have weaselled their way into discord's moderation team to label everything even remotely sexy as being the most egregious thing they can.
Flooding the database would (in their perspective) have the added bonus of just removing ALL mature content from the internet, legal or otherwise.
E
Is there any proof of this? Sorry not being rude I just want to know if they're really like that before I label in my mind, what seems to be a good cause, as bad
Well, they're website is pretty obnoxious about their stance on the sex industry
Everyone who sees this should look into Exodus Cry and everything they've done, they have more power than you'd think.
@@armchairgeneral7363it's exploitation. Don't legitimise it by calling it an "industry"
Discord seems to be getting worse over time, its just, implementing new things, that make new problems, then it turns into a cycle. But great video!
@Yan燕爱 no
@KandersV36 ok wont watch him anyways
@KandersV36 "Joined 43 minutes ago"
This is why the FCC needs to step in and only allow AI to assist moderation if a real person makes the final decision.
How would this be a FCC point of purview?
@@retchedThe FTC can sue tech companies to make changes
@@MaahirMomtaz12 Precisely what I was implying.
Sadly the FTC now has extremely little power
Yeah, but we're moving into 4 years of "gov bad, corporation good" so good luck to us all
Discord Child Protection Logic:
"If everybody is a Predator, noone is"
unjust and ignoble worlds logic: 'if you do this, you get ban'
just and noble worlds logic: 'if you do this and it can be reversed and removed, you have a [timeframe] to remove it, othervise face a ban in the face'
5:49 Imagine the mod that was forced to watch that just f*cking hates Marvel Rivals and decided to flag it as CP just out of spite LMAO. Wouldn't even put it past them tbh.
they're more of an overwatch stan than a taylor swift stan is to the popstar lol
Based. Fuck that game.
Who asked? My content is superior.
please don't give any attention to that attention seeker above me.
@@NihigetheQueenLOL if it was you would have 1.18 million subscribers
AI needs serious regulation.
*bribe* FTC and FCC says AI dont need regulation xD
Craziest thing about the appeals not working is that a week ago I got someone banned for the same reason, and they appealed it and got their account back.
I won't be surprised if there will be an entire AI based discord server that got self reported by itself (AI), and then got banned the very next day with this reason: *AI GENERATED CONTENT*
What@Yan燕爱
@Yan燕爱 can we get at the part where you cut me face with giant scissors
@@iulius2226I REMEMBER THESE GATCHA STORIES
@JonniShandorIt’s an urban legend.
who cares; my content is butter. 😂
I think it's because in the bottom right of the video it says "teen," so this AI saw inappropriate video and then saw the word "teen" and assumed it was a teenager in the video. That's just my guess :,)
they should know its a video game...
@@itzryan_1 If it's a new character it has no information or before reference to tell it's from a video game. This is AI, after all :)
@@SyreDoesThings oh mb
@@SyreDoesThings The AI could, I don't know.. check the accompanying text...?
Stop covering for corpos.
@ Hello Dakota, I never once stated I was covering for corporations. I was actually saying their AI is very dumb and associated the age rating for the video game to be related to the contents of the video, as that's my guess. That is clearly stupid because it's very recognizable that the AI was wrong, hence the whole point of this video. The AI is and was not capable of checking accompanying text because it is dumb and a cheap option for moderation, which is why this happened in the first place. The AI made a mistake because the corporation messed up on checking it's faults.
As a human moderator and community manager of many years, I can definitely say that computers won’t be doing our job for years. It just doesn’t do it right and AI will even admit when I have conversations with it that it doesn’t understand nuance. It sees boobs and thinks oh no, not safe for work… but the context of that is that the woman was wearing a dress and it was on the floor of the Emmy‘s… Yeah, not NSFW rip AI rip
AI can only do so much and it is stupidly quick how much diminishing returns you get with AI-based systems. This is because the technology is iterative since it enforces itself with more data but then it suffers the same issues as generative AI where it begins to "hallucinate" and flag patterns so generally that anything even remotely similar would trigger a report. AI currently is pretty good at recognizing something very specific but anything more general it started to fail because it lacks any higher level processing capability (as in, actual intelligence). All companies will very soon realize that limitations to AI systems will make them lose productivity so they will be begging people to come work for them again.
@@TheOz91 Begging them to work for moderation?
They would probably need to hire hundreds of thousands of people to do so, and even then they would need high pay because they have to look at CSAM and inappropriate content
AI moderation is not getting replaced with how large these kinds of platforms have become, however, there might be soon a breakthrough on machine learning for specifically detecting inappropriate content, that is only if these companies invest tons of resources towards figuring out how to do that, and they obviously won't unless the law pressures them
@@silly.spongus Sure but the tech is still going to be inferior compared to people. For that reason, they will still need to have human "supervisors" because despite the progress of ML and AI in general, a lot of false positives will still exist. Again, AI will never be truly intelligent. I do support the use of AI in these sort of jobs but just like every job that companies are trying to replace people with, the diminishing returns will force them to hire back at least a few skilled staff members.
@@silly.spongus They make enough money selling your data and from idiots giving them money from subscriptions. The least they could do is pay *real people* money to make their shit less dystopian.
The AI probably flagged it because of the T for Teen symbol in the bottom right.
How does this make any sense
@@gairisiuil Ask the AI, cuz idk either
@@gairisiuilthe AI and human probably thought the character was a minor based on the "T for teen" thing in the bottom right most likely no lmfao. yet somehow real crap sneaks through and doesn't get flagged
They were already using AI over a year ago, I got hit with a ban from their AI over a false match over a year ago & it took like 2 months to finally get a real person to review it & get my account back. It took way too long to get an actual person to look into it, I've not resubscribed to nitro since, I still use discord but I have no trust in it anymore. When a AI can instantly perma ban you for something like that & it takes months of spamming support to get a real reply that is a big issue.
*sigh*. Discord.
@KandersV36💀🙏
@KandersV36 Fun fact:
You're a bot!
Bot account speedrun any%
@KandersV36you sure 😂
@KandersV36shut up
5:09 A human reviewed my appeal on the copyright strike on my video about a song from Mojang's creator safe playlist in a Minecraft video. Within 5 seconds they had received my appeal, reviewed the ~2minute clip and answered, as a supposed "human".
UA-cam is 100% getting sued for their ai shit.
Is it possible that instead of Discord picking up this post, flagging it, and sending it to the NCMES, that maybe another company did this and when it hit the NCMES database Discord AI acted on it? The clip of the website shown in this video shows that "once items hit the database all other applications can see it" (paraphrased, from memory)
Not _impossible_ , but it likely would have come up in other forums as well... Marvel is, I'm pretty sure, not a terribly small thing, and if X(Previously Twitter) _isn't_ using AI moderation... We'd likely have seen some fallout from that as well.
ai will not be fully replacing moderation anytime soon
edit:if you see ANY bots report them for spam
edit 2:is ntts or youtube eating the bot replies idk but its good either way
@Yan no, no you’re not
Roblox moderation is already entirely replaced by ai so that's not entirely true.
@Metal-Sonic Roblox is halfway to realising the dead internet theory.
you mean in quality right?
Because clearly, it has and will continue to.
@@Tagiau yeah it's literally just a race between Discord and Roblox to see which brings the dead internet theory to reality first.
I know someone how got suspended because he joined a server which had a secret channel that he didn't have access to and someone posted a guide to make a bomb in that channel. Discord basically banned everyone on that server not just the on that sent it
As someone who does moderation myself, AI should never be in charge of actually placing bans / infractions. The most they should ever do is flag to a queue for humans to look over.
2:21 I am actually impressed that you can somehow talk to a woman for 12 minutes. That's a huge accomplishment
That doesn't necessarily mean 12 minutes with one woman, it could be 1 minute for 12 different women, or even 1 minute but 12 different times with 1 woman
Situations like these remind me to not put all my eggs in one basket and not to put my communication methods in the hands of one company. This goes for other tech things as well, use multiple operating systems or at least know how to and keep a copy just in case something happens to one.
There's Matrix (Element/Fluffy/etc.), TeamSpeak and Revolt Chat. But a lot of Discord users aren't aware of better alternatives.
I’m sorry Rain world… here of all places 0:32 I NEVER see anything about Rain World when it’s not just the game but I see it in the “Discord ai is banning people for no reason video” ah yes makes perfect sense
is that the living pipebomb from rain worlb
Rainworld mentioned!!!
Lol wth is arti doing here?
Wawa mentioned release the 14 king vultures
That was my reaction too I was just about to write a comment about that 🎉🎉🎉❤❤❤
Free-use ai was a mistake.
@Letaryu cleverbot esp, nerdcubed did a video on it way back and one of the things it said was "*gives master water*" and that was uncomfortable then and it's uncomfortable now, considering how cleverbot works
Still sounds better than AI monopoly tbh
@Letaryu This comment reeks of someone that has no idea how AI works lmfao. Siri has never been AI.
Ai was a mistake
@Letaryu brother im in my 20's
Good old AI hallucinations.
basically that whole guy eating chips thing again
That gif where the guy eats popcorn was ACTUALLY from a CSAM video, so that mass ban made way more sense. Sue Storm in Marvel Rivals is canonically in her 60s and used technology to make herself look young. She's literally the oldest character in the game. 💀 It's not CSAM. 💀
@@KrimsonKattYT how is that from a CSAM video
@@naveednawwaf It was clipped from that 💀
@@fujiwaranonekobiodrando1257 Theres no way this guy is originally from... this kind of video 💀
So what I'm hearing is, if someone inserts a popular gif or other non-offending image into illegal materials and that gets pickup by AI, hilarity ensues
I actually had an old child safety ban overturned after going to court over it. What happened was that a server I was in got raided, and the raiders posted and reported the stuff they posted. It was so much that it couldn't be cleaned up in time and the entire server got nuked. Discord, in court, tried to argue that I was responsible for it as I was in the server when it happened. Judge ruled in my favor, and we settled for damages and an unban. Discords actions have always been questionable, but it's the first time it got stuff wrong on such a scale. Hope they will fix the system and especially allow appeals in the future, because the default no appeal possible is just not good enough.
I got banned one year ago for being admin on a banned serwer. It was banned because of child safety with no further explanation. This was a discord on making desserts. Mostly people posting pictures of cookies they made and discussing what the best brownie recipe. I have no freaking idea why it got banned.
Ngl the AI probably saw a sexualized character and the "TEEN" rating and was like "yup, its CP alright"
I hate it when this happens, now a new excuse for creeps try to get away with it.
Fortunately, Discord did send out the emails with free nitro gifts as an apology for the mistake, which means Discord did take up responsibility. Discord sending out those emails caused a huge copypasta about Discord's email system got compromised or a massive phishing attack impersonating Discord.
Damn. Now we gotta trick the scammers into doing this
But i did 😌@Yan燕爱
It's patched
@@Shavix i don't think they patched all of fake csam triggers. Just gotta find them smh and trick scammers
AI should flag the bad content, then send it to a human to make sure its legit.
you're right actually
This is why I'm not using the forward button.
Imagine if NCMEC sued companies using AI submissions for literally endangering children by making real CSAM harder for them to track.
Stuff like this keeps happening yet companies STILL insist on using ai. Mindboggling.
money saved on moderation outweighs money lost from banning nitro users
I got sent actual CP in my dms yesterday and then I reported it, guy didn’t get banned and now people are getting banned for forwarding “CP” even though it’s just a normal image.
How the hell do you get all the interesting reportable dms and i get like, nothing, man i wanna report people bruh!!
Same. A dude that showed me his litteral files of pics and videos. (He was edgy, telling me he liked to study the anatomy. Like bruh)
He also didn't get banned.
I can't tell what it takes to get a user banned for actually committing a cime instead of being thirsty for a Marvel Rivals skin. It's truly pathetic.
@ yeah its stupid
that's awful man.. did you report the person to the police at least?
A friend of mine was caught in the Germany bans and it took him over a month to get his account back and only got it back, after I talked to a friend who knows a Discord employee...
the maid one is back
Fun fact UA-cam's moderation is shit
Heads up, this commenter is a bot, it joined 18 minutes ago as of writing this comment. (Kandresv36)
?
the kand spam bot's profile also links to... the youtube le troll squad 😭 (i don't think i need to explain who they are)
Hello fellow Arcane fan @@cathylin6140
Discord banned me, said “You broke the rules we won’t unban you” then unbanned me immediately
The whole NCMEC getting flooded with this stuff is why people reporting lolicon to the NCMEC despite them literally telling people to stop. Animations are not missing or endangered children. Don't do that. Remember that every time you report lolicon to them, you are actively responsible for a child being hurt.
Dude, my partner had their discord account permanently suspended for this... we were freaking out, had no idea what was wrong, years of memories lost lol.
Don't use apps with censorship. Nevertheless it's disgraceful.
Go make new memories by going outside and touching grass
create own system of operating, own sites and own nation and history and declare yourself as above all human bc you have a righterous morality superior than other people's minds
@@kaydog890 don't know if you're just stupid or genuinely don't know because you don't touch grass yourself, but people can use discord as a messaging application AND have a life of going outside and touching grass at the same time. maybe don't be fucking stupid next time you comment, hope you get better
@ Speaking yappanese
There’s so many disgusting people on discord not getting banned, but someone forwards a game update video…. That’s what gets banned
thank you for not calling CP something stupid like "cheese pizza" or whatever
He actually did call it cheese pizza in another video
Edit: Video it's said in is "PSA: Don't send this Photo on Discord!"
How about dont shorten it and just say it
@ cus that’d absolutely get you demonetized, cp is a good way to refer to it safely without giving it a much lighter tone than it deserves
@@fujiwaranonekobiodrando1257 Most likely because the actual term can't be said in a video because of monetary reasons? I'm not entirely sure but it's why a lot of channels have to tread lightly because they know UA-cam is sensitive about certain topics but like I said I don't entirely know for sure
@Retro_Lightning It's really weird because cp is shorten for a lot other things here :(
i talk to the same 5 people every day, i didn't even open the marvel rivals server even tho i am in it, didn't share it and still got banned. like... come on...
Sounds like Discord may be flagging people in that server as a whole. Or there’s someone mass flagging accounts in that server.
amazing, got banned for existing
Good thing, now you're free
It's funny, shitty AI checking answers from shitty AI is literally a strategy to make AI more competent called "Chain of Thought"
If you ever think Discord Bots are stupid, Remember the one time a bot gave me a warning for just sending an "XD".
Judging by the photos of discord that are on Glassdoor, this doesn't surprise me.
Ironically I had the same thing happen (not for the Rivals thing) but as part of the AI sweeping a bunch of accounts. I recommend messaging them on X (both Discord and Discord Support) because there’s at least one actual person who checks it. I’m saying this if you’ve already gone through the other means. I got my account reinstated, and the issues fixed. Just took them like a week and fights with the Clyde bot. But yeah, I wouldn’t be surprised if it’s just AI at this point. Their system is so broken.
I can't believe they somehow managed to detect an entire language as spam
Discord AI Situation Is Insane
Those who know
good dayy for the unemployed
@Yan燕爱 you’d prolly look good in a Low Taper Fade
6:38. Roblox had a similar issue and these two major issues are actually partially intertwined..... Its disgusting and sad
i hate the false roblox voice chat reports
@splunderman youre probably guilty of actually saying bad things.
Roblox shouldn't have ever added voice chat
in roblox never say a word and you will be safe
in discord never post anything and you will be safe
and i thought microsoft will ruin discord no they ruin themselves
10:03 discord basically just called bro a pdf for making a discord account
It took me about five minutes to realize that you didn’t mean PDF as in the file type..
@ lol yea i havent tried but im pretty sure youtube doesnt like it when you say the actual word
ai is like an eight grader in some things but also its acting that it cant do 2+2 and 2×2
It can't differentiate can from can't and doesn't know him many R's are in strawberry
@KandersV36shut up we know it is not your videos, we aren’t 5 year old brain rotted kids
@KandersV36 🤖
plz learn english better to not get bullied@@3enn7
Discord's broken, we've established that many times now.
0:38 Artificer Rain World! :D
Artificer got banned on discord for hating scavs so sad
rain world is peak
now i hate AI above everything cause companies use it as the worst excuse possible and con artist use it as their golden ticket
but AI in video game is completely different and rain world AI is one of the best i seen that actually
until i get on raindeer and that thing is the reason why the grass stops a run
She was spreading anti-scav propaganda
@@Zenala OOMF?????
The voice over at the end was diabolical ( I love it )
man had to bring up breaking child saftey, i some how broke it once and i dont even know why, then he has to say its the worst thing to be flagged for :((.... anyways great video
They still havent fix the darn false banwave before this one still
I actually got banned for the Marvel Rivals incident forwarding the message with my comment of “the snowbunny propaganda grows”. Thankfully Discord reverted it shortly and gave me three months of nitro for compensation
Legitimately thought it was because of a free subscription I attached to a prepaid card that had like a dollar on it. Nice to know it wasn’t that IG
Any web service using automated filtering needs a class action lawsuit for being lazy.
AI makes sense for mass reports, however AI should only handle the obvious, but there should be a team of people to double check reports that the AI might be unsure of.
I have no idea about this case specifically, and my ban was not involved in it. However, I find it amusing that this banning spree happened THE SAME DAY I got falsely banned out of nowhere. I was just minding my business, and then I got forcibly logged out and said my account was permanently suspended for "financial scam," whatever tf that implies. I got my account back, but it was such a headache to deal with because my thought process is that a hacker who scammed me months ago got into my account at *some point* between that scam and now and got reported. Thankfully, discord responded to me after countless bot responses (unlike they did at the time I was asking for help from that scam) but it should have never happened in the first place.
I'm Curious how this will turn out?
@KandersV36 me when i lie
^^^ bot above me ^^^
I'm on the verge of writing a script to mass-delete every message I send about a month after I sent it, just to prevent some random AI from looking at it and deciding that I should cease to exist.
Lol, that doesn't actually get rid of it. If servers can see what you deleted on the mod log, so can Discord
@@richardmillhousenixon the default mod log servers have doesn't actually save messages people deleted, only who deleted them and when (seriously that's incredibly stupid for a multitude of reasons, mod logs shouldn't save deleted messages).
The mod logs you see are all done using third party bots that just repost what people send after it's deleted.
Still, discord does probably keep your messages for a time even after you deleted them (it's a waste of space to keep them forever, Discord is already bleeding enough money as is), so all deleting really does is save you from someone reporting your old messages.
if you read their tos, you'll find out that they keep all your messages for like half a year after you delete them
NTTS, moderation AI predates generative AI and I'm pretty sure Discord and other platforms has been using AI moderation for at least awhile now, platform like UA-cam probably a decade by now. Machine learning and algorithms have been used for many things for many decades.
Unless you know they've changed that up their moderation system significantly, we shouldn't assume beyond what we can already surmise. This new prob could easily be a bug, but also humans have made errors like this too.
I feel as if you're unfairly blaming AI, AI that you don't even know exists within Discord beyond the industry standard algorithmic automated moderation filtering. It's the same old cycle, we bIame the new unknown and spooky tech for problems of yesterday. We just keep shifting who's fauIt it is.
I am glad I never forward anything, or even saw this announcement. I would hate to lose my account. I have had it for years.
"cause the want to save money"
Me (gets a liberty mutual ad right after that) 😂
And this is why I also use things like Matrix and IRC as a back up.
Ch!na is not gonna be happy about this. They own a large part of discord and they made Marvel Rivals.
You Cant even write something China - Unfriendly in the chat
discord not banning actual CSAM and banning people for THIS is so dumb. this does NOT need AI, there's ways to check image hashes automatically and from what I can recall, there's services that provide those hashes to filter out CSAM. AI should be a last resort with how unreliable it is, and even then, as that one user in the video said, there should be a good way to treat the false positives.
bro they banned me for ban evasion 💀
Same bro
A better approach is use AI to guess whether it is offending, THEN send it to a human moderator, they'll determine if they get banned and then they can even train the AI to be better.
Wow dude your upload schedule is peak
No @Yan燕爱
@Yan燕爱 your account is 4 days old and I'm the third person on this video which you ask this question
@@camtoorealthey are probably a bot
One day discord is gonna do something people will sue them about I'm just waiting till that happens.
I hope people charged back all the money they ever spent on nitro and cat ear avatars.
If I'm not wrong, many big social media platforms switched to using AI to detect child pornography as it is less controversial than a human looking at what potentially could be child pornography.
when pointless stuff turns out to be more pointless than ever
This is why I always take screenshots of my friends usernames, and the servers I’m in, can’t trust discord
Bots spamming on replys on a video about ai? Who could have seen that coming. (Or some loser in a basement, still probably a bot)
Yeah I refuse to pay for nitro. I'm not gonna give a company money when I don't even know if my account is safe there.
1:12 *extremely loud incorrect buzzer*
That "free nitro gift links" email being real is absolutely wild -- people are absolutely going to send phishing emails that look exactly like that one now.
3 of my german friends got banned from discord mid call shit was so confusing
Sue storm is not even a child💀
historians will skip our generation 💔🙏
I would much rather pay money to a service that has ACTUAL HUMANS running moderation of said service than a service that uses AI that EASILY gets abused by some of the worst people on the planet. Facebook, UA-cam, Twitter, Discord, every social media platform, every single website or application that uses AI as a major source of moderation all deserve to lose a shit ton of money for that reason alone. If I have to pay more for humans to make a good service, EVEN IF SOME THINGS TAKE LONGER, I'll fucking do it. When I pay money for a service, I expect PEOPLE to be moderating and handling the service. NOT an AI backend.