Forgot your password?
typodupeerror
Privacy Your Rights Online

Discord Tries To Walk Back Age Verification Panic, Says Most Users Won't Need Face Scans 123

Discord has moved to calm a user backlash over its upcoming age verification mandate by clarifying that the "vast majority" of people will never be asked to confirm their age through a face scan or government ID.

The platform said it will instead rely on an internal "age prediction" model that draws on account information, device and activity data, and behavioral patterns across its communities to estimate whether someone is an adult. Users whose age the model cannot confidently determine will still need to submit a video selfie or ID.

Those not verified as adults or identified as under 18 will be placed in a "teen-appropriate" experience that blocks access to age-restricted servers and channels. The clarification came after users threatened to leave the platform and cancel Nitro subscriptions, and after a third-party vendor used by Discord for age verification suffered a data breach last year that exposed user information and a small number of uploaded ID cards.
This discussion has been archived. No new comments can be posted.

Discord Tries To Walk Back Age Verification Panic, Says Most Users Won't Need Face Scans

Comments Filter:
  • by NotEmmanuelGoldstein ( 6423622 ) on Wednesday February 11, 2026 @05:28AM (#65981848)
    The reality is, "know your customer" and "think of the children" causes identity theft: The effect upon criminals is minimal. Because it's moved criminal intent from an individual cheating the system, to a corporation of serfs moving illegal goods.

    Discord is a target of 'child safety' spyware demanded by the governments of the world, and a target of users who realize Discord tracking them, makes them a victim.

    • by zlives ( 2009072 )

      perhaphs a pki-like verification system that a "trusted" 3rd party verifies age without giving id/personal data to a known leaker,

      • This is by far the easiest. One goes to an age clearinghouse, get a certification so they can link an account, even if it is their troll shit-post account... who cares... it just shows that the account is over 18.

        However... dataz. Being able to sell to age groups is a big thing, because kids buy a lot of stuff, especially at the vulnerable times where imprinting them with needing to buy items or wear products in order to be accepted into daily society is a must. Everyone wants that age data, either to se

      • In mainland EU we task government with creating ID. Just let them sign a NFC passport which I can scan with an app to certify my age and give the website a pseudonomous identity. With a CRL instead of online certificate check, and no way for government to connect pseudonym to proper identity.

      • Comment removed based on user account deletion
  • ... too complex bloatware for instant messaging wants personal information? i smell big brother, and since i am irl big brother, one is enough.

  • by devslash0 ( 4203435 ) on Wednesday February 11, 2026 @05:49AM (#65981860)

    Most countries that introduced age verification requirements state that verification must be "highly effective". Would this predictive profiling be enough in the court of law?

    • by AmiMoJo ( 196126 )

      It's probably stuff like the age of the account. If it's 10 years old, the chance that the owner is under 13 and registered it as a toddler is quite small. If they have made purchases with a credit card, or linked a phone number to it, that sort of thing.

      • It's probably stuff like the age of the account. If it's 10 years old, the chance that the owner is under 13 and registered it as a toddler is quite small.

        Can someone please tell this to Ebay.
        They regularly sent me emails in which they thank me for being a user for more than 21 years, but request I use a credit card to verify my ID to prove my age if I am trying to buy a tool with a sharp edge. (including a pair of scissors with a blade less than 1" long).

        I live in the UK, and do not have a credit ca

        • It's probably stuff like the age of the account. If it's 10 years old, the chance that the owner is under 13 and registered it as a toddler is quite small.

          Can someone please tell this to Ebay. They regularly sent me emails in which they thank me for being a user for more than 21 years, but request I use a credit card to verify my ID to prove my age if I am trying to buy a tool with a sharp edge. (including a pair of scissors with a blade less than 1" long).

          I live in the UK, and do not have a credit card. I do have several debit cards. Ebay does not seem to understand that some parts of the world are not in America.

          You had me until you said you live in the UK. As a Canadian, it pains me to say, but the UK is even stricter than Canada on this surveillance and violence prevention crap. Even if Ebay didn't demand you prove your age for wanting a pair of scissors with less than an inch long blade, the UK Post would likely demand ID before delivering the 'dangerous goods', to track the movement of 'nefarious instruments'... lol

          In case my sarcasm wasn't evident. I feel your plight, and share in it. This nonsense needs

      • by gweihir ( 88907 )

        Level of language, types of topics, age of account, etc.

      • Comment removed based on user account deletion
    • We've covered the accuracy of predictive profiling time and time again. The amount of insight you can gain from someone's behaviour is truly immense. Remember the case of Target being criticised for sending pregnancy related products to a teen girl based purely on purchasing behaviour, only for it later to come out that she was pregnant despite her not telling anyone?

      Highly effective is a word distinct from "perfect". I think it would be an easy bar to pass algorithmically given the information platforms sl

      • by zlives ( 2009072 )

        i am not a discord user, does discord monitor the app usage/conversations are not private? i thought the servers were private enclaves without giving access to discord the company?

      • by Luckyo ( 1726890 )

        If I remember correctly, the pregnancy thing went further. She didn't even know she was pregnant yet. The system just flagged her behavioral changes as associated with pregnancy.

        Because most of our behavioral patterns never reach conscious mind. They're decided on a different cognitive level. Our awareness of ourselves is quite low for evolutionary reasons.

        • It is a made up story. It literally never happened and was. fabrication
          https://www.predictiveanalytic... [predictive...sworld.com]

          Slashdot disappoints me because news that a teenage girl got pregnancy junkmail because of a robot superbrain should have seemed suspicious to ALL OF YOU. Why would the father go to the press with that shit. Come on it was in every news outlet like boom within a week. Why would target admit to doing such a thing to a teenager instead of just saying it was a mistake?

          So easily influenced.

          • I want to make a comment about you using an unheard of source for your story rather than any official retraction, and a source that (I cannot believe anyone would be so stupid) copies not just the name but also the iconography from well known hack tabloid outlets rather than anything legit. The article itself postulates nothing but speculation and ties the result to selling of a book.

            I think I'm far less easily influenced than you are. Go look up an authorative source on the topic. I'm keen to know I (and m

            • nobody in the story behaves like a real person. the dad, target, and the pregnant teen.... all their behavior is wrong.

              I'm personally embarrassed i had to see the story in the press a few times before i started wondering and looking for a source.

            • I have to confess i never even read the story I linked. I simply trusted that says everyone in the story acts wrong and that nobody can explain how a teen's very private pregnancy became a viral news story.

              Target wouldn't want a crying teenage talking about the dangers of big data... unless there is no teenage girl, then they wouldn't be scared of backlash they can't contain.

              The father is certainly not going to go to the press that his kid is knocked up and he figured it out by paying close attention to t

      • Remember the case of Target being criticised for sending pregnancy related products to a teen girl based purely on purchasing behaviour, only for it later to come out that she was pregnant despite her not telling anyone?

        This story is just propaganda and never happened.

        • Then you can post examples of retractions in the news? Because it was covered by reputable news sites which do in fact post retractions.

          • They don't post retractions for this sort of thing jesus. It's like you've never once identified bad reporting in your entire life.
            Personally i thought you were a old professional so you should have, at this point in your life, been involved in something misreported in the news

    • by allo ( 1728082 )

      Other laws also have weasel words. You are not allowed to circumvent an effective copy protection. Is it effective, if you can circumvent it?

    • If this was targeted at actual legal requirements then this would be a question to ask. And if age verification measures were actually about age verification then again this would be a question. The "save the children" laws were never about age verification. They are about the lucrative sale of information between five-eyes and the wider intelligence sharing.

      Five eyes et al are about getting around national laws forbidding surveillance on own citizens by getting other countries or organizations to do it

  • by Qbertino ( 265505 ) <moiraNO@SPAMmodparlor.com> on Wednesday February 11, 2026 @05:54AM (#65981864)

    ... than yet another glorified and way overrated IRC/XChat clone like Slack or other candidates - will only take me a few minutes. As soon as they get more annoying than convenient I'm out. For all I care they might as well just go broke already, since they're still not profitable. No need to try to gain any mass by starting this uber-surveillance thing. It won't pan out anyway, so they might aswell just give up.

    • by radarskiy ( 2874255 ) on Wednesday February 11, 2026 @09:31AM (#65982104)

      It's easy to replace Discord if no one wants to talk to yoiu.

      • by Anonymous Coward

        thanks for the unnecessary trolling today. we all really appreciate it!!

    • by Anonymous Coward
      Because none of the replacements will ever want to comply with the laws or any weird stuff like that. You'll be safe forever !!
    • Besides sibling comments' objections, I don't know literally anyone who uses Discord on purpose. People use it because they have to in order to get support with a product or service, because it's where their gaming clan meets, etc.

      • 'The people' aren't choosing Discord, it's the forum creators that want control over a private fiefdom - and apparently Discord makes this easy for the creator.
        • by G00F ( 241765 )

          because websites are hard and apparently passe... so they turned to the worst wordpress platform there is.

          GOD! I hate discord, but that's where users are, and have been trying to nudge people to matrix, which is still a pile of crap but at least allows open communication.

          I miss icq/aim/yahoo messenger.(oh the days of using everybody and pidgeon!!) I miss being able top send files to friends. (and no, things could work even when both parties nat'ed) and things can still be open and secure even if something

      • we got pulled into it by young normie gamers who we should have used bullying and peer pressure to get them to install mIRC.
        if they start talking about emojis and scrollback support we should talk in funny voices and imply their tastes are unsophisticated and they're personally dumb and ugly.

  • by Zarhan ( 415465 ) on Wednesday February 11, 2026 @06:04AM (#65981870)

    My daughter plays Roblox. They are asking her to take a selfie to get assigned to an age group.

    https://en.help.roblox.com/hc/... [roblox.com]

    Seriously, they are planning to use some AI doodad to place you into a group "Your estimated age helps place you in the appropriate age group (5-8, 9-12, 13-15, 16-17, 18-20, and 21+) to customize your experience on Roblox". I've known people that have baby faces in their 30's, and I've seen preteens that have started showing off beard. And this is all based on some image recognition thingamajick?

    Most countries at least in EU have some sort of governmental identity mechanism, that allows you to make strong identification to any service. Why does this have to be some sort of AI magic that steals your data, when this all could be handled by a simple federated identity tokens. Just go to your government site, ask it to generate a signed token that states "On this day, holder of this token is X years of age", lifetime of token can be like 15 minutes. No need for government to learn what were you visiting and no need for the service to know all other details either.

    They are already utilizing some third party to do the selfie recognition and evaluation, why couldn't the same third party pass a token between gov website and the service with the same price?

    • by allo ( 1728082 ) on Wednesday February 11, 2026 @06:31AM (#65981888)

      No method (regardless how clever) and solve the root problem: A child can ask its 18 year old friend to unlock the account.
      You can find ways to prove age, but there are no reasonable ways to prove the current user is the person who proved to be 18+

      • I'm going to require every user of my platform to submit a brethalyzer sample after taking a whiskey shot and burning down a Marlboro Red.

    • The logical thing to do, and which can be done even by a savvy child, is to produce an AI picture and submit it. Odds are that the system won't detect that.

      I had a full amish looking beard when I was 14 (my moustache grew in last) so people had me buy them smokes.

    • It's always about monetization becuase big tech is just ad tech and marketing tech. They've got KPIs to meet.

  • At least the video-thing. They stated processing locally only and nothing about storage. That means, at least in Europe, they are not legally allowed to do any storage and they must keep the data local or being in direct, non-accidental violation of the GDPR. Fines go up to 5% of annual global (!) turnover and, for repeat offenders, a prohibition to process or store any personal data.

    Now, I have zero trust in their integrity or honor. But people will be looking and exfiltrating that video data or storing it

    • Re: (Score:2, Informative)

      by Anonymous Coward

      They already had a leak of data they were supposed to have deleted. I wouldn't trust them.

  • Now every user needs to think about if they behave like an underage user.
    Better not login at times pupils have free time. Login early on holiday days to prove that you're not a child sleeping long at the day off. Make sure not to click like on silly memes. Join some servers that look boring to kids.
    That's just another way to make users self-censor.

    • by Calydor ( 739835 )

      You know, when you put it like that I realize Discord might be massively improved if users are given incentive not to act like they're underage and immature.

    • by gweihir ( 88907 )

      You just ask for age verification. At that time, if they find you to not be underage, they do not ask again. As reverse ageing is not a thing, that makes sense.

      • by allo ( 1728082 )

        Maybe I skipped part of the thought. I suppose that the behavioral part to suppress the verification works reversed. If you join some Roblox Discord, make a lot of typos, and are inactive during school hours, I suppose that Discord flags your account and asks you then for verification. I won' rely one some adult pattern setting the is_adult flag permanently, but I would fear that once you triggered may_be_child they won't let you go without verification.

        • by gweihir ( 88907 )

          Well, in the EU they need permission to do the profiling. If they get permission to do it once, they cannot continue to do it legally. While technically they could have some anomaly detection on verified adult accounts, I doubt they do. Remember they do not actually care, they are forced to do this.

  • According to reddit, users are approaching this by:

    1) researching fake ID

    2) https://github.com/stoatchat [github.com]

    3) Matrix / Element

    4) Mumble

    5) Some going to Slack, apparently

  • by heldal ( 2015350 ) on Wednesday February 11, 2026 @08:01AM (#65981940)
    At last, we get a Fix for the disease that has threatened to ruin the internet. Don't you see? By using behavioural profiling, all the immature idiots, regardless of biological age, will be either forced out or grouped together with similarily minded hoodlums!
    • You're probably not serious but just in case, anonymity is an important condition for free self-expression. Many things I write online, I would not want them tracked to my sorry self. Slashdot is a good example on how an anonymous forum can stay polite and civil (mostly), without requiring proof of age or ID. If we can do it, why can't the others?

      • they're pretending to protect kids and this is really about putting identity within easy reach of the system whenever someone powerful needs to put boots on faces.

  • by Anonymous Coward

    Don't be a nanny about it. Let the parents decide what's appropriate for their kids.

    It should be none of the accounts that are asked to verify. Don't try to predict the age either. Nobody wants this

  • So, tell me.... (Score:2, Insightful)

    why the fuck doesn't someone pursue criminal charges of neglect against parents for allowing their children unsupervised access to a known dangerous place (i.e., the internet).

    WTF happened to parental responsibility?

    Their damned kids would be safe if they were properly supervised.

    • by Anonymous Coward
      2/10 troll
      Not even MAGAs are stupid enough to fall for that
      • Re: (Score:1, Troll)

        What? Fall for actually looking after their children? Fall for supervising their activities when on the internet? Fall for looking out for their safety?

        Please, do explain your stance.

        You're playing a fucking posting game.

        I'm saying parents need to supervise their children on the internet.

        Apparently, you're firmly in Epstein's camp. Fucking pedo.

    • Half of all parents would be in jail. It's an epidemic - mostly because both parents are too busy putting food on the table because of the huge wage imbalance that keeps growing.
  • My server is at least 10 years old, probably 15. So unless they think I started it when I was 3 years old, they better lay off. Also, not an NSFW channel. So if they come asking, they're getting a pic of my balls.
  • by HnT ( 306652 ) on Wednesday February 11, 2026 @09:55AM (#65982166)

    We went from privacy consciousness and being scared of government bugs planted everywhere to giving ALL our information away for practically free AND running as many bugs and cameras in and around our houses, voluntarily.

    It is absolutely more than enough to slap an extensive form or a prompt in front of somebody and they will happily enter all their information. Even games are now allegedly-AI-scanning your face to create your in-game avatar and people actually do this. Then they film and publish their lives and opinions, voice face and all, online and anyone who wants can use all that information.

    It is seriously bewildering all the data people are ready to diligently hand over nowadays.
    The backlash against discord is almost a surprise, it seems their reputation was even worse than the other data hogs.

    • It is seriously bewildering all the data people are ready to diligently hand over nowadays.

      "Why do they trust me? Because they're stupid f*cks." I think Meta's founder's quote serves here nicely.

  • Just for the "less government regulation" arguers......How did "too much regulation" cause this and how will "less regulation" fix this?
  • deadbeats (Score:4, Insightful)

    by Dragonseye ( 1103251 ) on Wednesday February 11, 2026 @10:09AM (#65982198)
    As far as I'm aware the only true way to stop children from accessing content you don't want them to see, turn the net off in the house or grow a pair and start parenting.
  • by Sloppy ( 14984 ) on Wednesday February 11, 2026 @10:39AM (#65982280) Homepage Journal

    IM needs to become a fully P2P (and non-proprietary) thing, so that there is no central authority (other than maybe DNS itself) to be coerced. Of course, that brings up NAT problems so that laymen are going to have connectivity problems, and a lack of any sort of commercial advertising and other pressures to get people to come over. (And I'm casually glossing over other problems, such as authentication.) No wonder it hasn't taken off yet.

    But if we can get there, the holy grail is a lack of any particular party being a provider of the IM service. We'd just have providers of the lower-level services (i.e. ISPs), who can't really interfere with individual applications unless encryption gets banned, which would mean no more e-commerce, so there should be strong lobbying (e.g. rich people like Bezos) on The Peoples' side.

    The goal is to have it such that whenever a law says users are required to avoid certain topics, it'll be the users themselves who are expected to comply with it. (e.g. If you discover that you are sometimes choosing to set the Evil Bit, then you are expected to report yourself.)

    Until we get to this point, IM cannot be secure or resistant to censorship (same thing).

  • I hope the backlash is severe and other companies take notice. Closed my account yesterday.
  • Create a smokescreen in the near term. Boil the frog slowly over a longer amount of time.

    Apply the ID changes to a few groups at first. Deploy it gradually to the rest of the groups.

    • by La Gris ( 531858 )

      Create a smokescreen in the near term. Boil the frog slowly over a longer amount of time.

      Apply the ID changes to a few groups at first. Deploy it gradually to the rest of the groups.

      This is the most likely outcome indeed.

  • They're not making it to their IPO.

  • Already deleted my account.

"When people are least sure, they are often most dogmatic." -- John Kenneth Galbraith

Working...