Monday morning saw an announcement by Discord of an upcoming change to their Terms of Service which should, by all rights, be infuriating to anybody old enough to buy beer. The company announced that starting next month, Discord would be engaging in a “phased global rollout” of age-verification measures for new and existing users. Discord’s head of product policy, Savannah Badalich, stated in a press release, “Rolling out teen-by-default settings globally builds on Discord’s existing safety architecture, giving teens strong protections while allowing verified adults flexibility.”
Naturally, none of this has any actual bearing on the safety of teenagers, or anybody else for that matter.
We’re From The Government (By Way Of The Corporate Office)
Discord touts in their press release that they “successfully” rolled out a teen-by-default experience in the UK and Australia last year. It should be noted that those two countries passed laws in the recent past requiring age-verification for social media platforms, porn sites, and other online forums which may contain explicit or “harmful” content (the UK’s Online Safety Act last August, Australia’s Online Safety Amendment Act in 2024). Their use of the term “age assurance” is a weasel phrase to try and make it sound like they’re trying to positively identify adults as adults instead of negatively identifying teens as adults. It’s utterly abhorrent either way, given than a responsible adult shouldn’t have to prove anything regarding their status as an adult and a teenager will inevitably find a way to circumvent whatever roadblock one throws up.
As the Electronic Frontier Foundation pointed out so cogently, the UK’s Online Safety Act is pretty much not interested in safety. The definitions of what constitutes “harmful” under the Act was demonstrated rather graphically when Reddit attempted to comply and ended up blocking subreddits relating to poker (doubtlessly a torturous inference to gambling), vexillology (almost certainly because no word with “ex” in it can be considered wholesome), and rickrolling (I’m surprised Rick Astley didn’t make a fuss about this). This was on top of other subreddits involving journalism, conflict reporting, public health, and LGBTQ+ related issues. It’s worse than mere censorship; it deliberately withholds information which is of public importance and is absolutely vital to maintaining an informed populace. Discord’s proclamation of their “successful” rollout in the UK and Australia not only sounds insincere, but takes on a decidedly craven and servile cast.
Up until now, credit cards have served as a sort of de facto age identification regimen. However, given the simple trick of sneaking a credit card out of an unattended wallet or purse just long enough to write down or memorize the number, its perceived utility has always been substantially overvalued. Long as nobody’s running the card for an actual purpose, it’s just a number with a name attached to it. But there’s another angle to consider: the lazy teenager. The kid who thinks digging into Mom’s purse for her Discover card is too much hassle when there’s doubtlessly free porn out there just one Google search away. In truth, credit cards are simply a barrier to porn with high production values and performers who get regularly tested. They do nothing to keep a teenager from stumbling down a rabbit hole into truly vile (and likely illegal) content purely by happenstance. Discord didn’t mention using credit cards as part of their efforts (which is probably wise of them). But should they consider the idea, it will likely not prove to be particularly helpful given the number of purchases which have undoubtedly happened for Nitro subscriptions and various stickers.
The Zone Of (Assumed) Safety
Discord’s press release mentions the “privacy protections” they intend to be using in their “age assurance” efforts. And as somebody who has been involved in the tech sector for a long damned time, and who’s been using computers probably longer than that, I can confidently say that what Discord is proposing is utter bunk, if not pure snake oil.
Their first claim is a “video selfie” which will be subject to “on-device processing.” The fundamental mechanic of the Internet is that a file gets copied from Source A to Target B. When you visit Game Observer, you’re downloading copies of the files which make up the page, the text, the images the CSS schemas. All of that is being copied from our server to your device. It’s why you have to clear your cache periodically. Somewhere in the process, some sort of information has to be going to Discord from your device. “On-device processing” suggests some kind of impending update to Discord which is effectively a machine-learning based facial recognition component, and we have absolutely no guarantee that such biometric data will not or cannot be exfiltrated (despite Discord’s assurances to the contrary), particularly since we have no idea of the criteria this invasive bit of AI is using to establish a baseline for a person’s age. Not to mention we have no guarantee that Discord won’t decide to “spot check” users in the future to make sure, “Yup, that there’s an adult.” Given the current capabilities of LLM-based platforms to create deepfakes of a person, anything which increases the possibility of such deeply personal data getting out into the wild should be considered anathema. The fact such data might be on be on your phone doesn’t help. The only way such data is safe is if it’s never collected in the first place.
Discord’s second claim is that “identity documents submitted to our vendor partners are deleted quickly.” So, assuming the “video selfie” doesn’t do the trick, you’ll be asked to provide, probably, photo ID. Which means even more sensitive information is being sent out, which Discord can piously wash their hands of should it turn out that, whoops!, that data didn’t get wiped off the partner’s servers after all. It’d still be awful if Discord handled this themselves. But given that the last “trusted partner” was subject to a data breach of 70,000 users’ government IDs (from the UK, as it turns out) back in October, any claim about how a third-party entity handles your data which Discord obtains from you should be given intense scrutiny. And probably should not be given any credence in the slightest. Discord’s updated talking points on Twitter the day after their original announcement keep harping on “trusted partners” without realizing (or at least acknowledging) “trusted” from a security standpoint means “the system or entity that can screw you over the hardest.” They’re asking their users to trust them to choose somebody who won’t screw up, and Discord does not have the track record to make that kind of ask without pushback.
Money Talks
Just after the new year, Bloomberg reported that Discord had confidentially filed paperwork with the U.S. Securities & Exchange Commission for an Initial Public Offering (IPO) with a tentative timeframe of March 2026. Which, by a staggering coincidence, is exactly when all of this stupidity is supposed to happen. Granted, there’s always time before the actual date for the company to decide that, no, this is not as good a time to take the company public as previously believed. They did the same thing last year. They’ve flirted with IPOs in the past and nothing ever came of them. But the timing seems just a little too neat in this instance.
It constantly baffles me how people will continue to throw good money after bad propping up tech companies which are in no way, shape, or form even remotely close to being a break-even proposition, let alone profitable. Discord has had venture capitalist funding pumped into it with disturbing regularity, always with an utterly ridiculous valuation attached to it. The last time that happened was 2021, when Discord raised $500 million USD on a valuation of $14.7 billion USD. Twenty-six years later, and nobody seems to remember how the dot-com crash happened.
If this is being done as an attempted tranquilizer for potential investors, Discord risks doing themselves a serious mischief. Part of their “charm” is the fact they’ve got a fairly sizeable installation base, over 200 million active users per month at last count and more than 650 million registered users total. Slash that number in half with defections over this particular policy, and suddenly Discord looks a lot less attractive. There’s no way it could pay back all those investors which have ponied up over the years in that sort of scenario. Discord would quickly become a target for private equity investors, who would then proceed to render the carcass for as much money as they could squeeze out before shutting everything down.
Cutting The (Dis)Cord
Let’s stop pretending that this is being done for the purposes of “safety.” It’s being done because moralist zealots and ideologically blinkered politicians think that it’s perfectly fine to spy on other people “for their own good.” It’s being done because Discord seems to be on the cusp of going public, and the fear of possible misuse for sexually related situations is somehow considered to be a bigger problem for investors than a decade of unprofitability. It’s being done because nobody seems to have the spine to resist the “insidious encroachment” Louis Brandeis warned about so many years ago, and those who do are fighting a desperate battle against impossible odds.
People crying and beating their breasts about “the children” are very much like those who are equally demonstrative about “the poor.” They’re faceless groups to these people. They don’t have any actual stake in the situation. They just want to make noise about it, and use those faceless groups as a bludgeon, all the while exhorting that they’re trying to make the world a better place. So certain are they in their rectitude, they completely fail to realize the damage they’re doing. And if they do, they rationalize it away, chalking it up as “too late now, but it’ll all work out in the end,” which it never does.
As a general rule, we want our kids to be safe. We want them to grow up happy, well-adjusted, and become functional members of society without any hangups or issues. But that safety does not come about from hiding them away from the world. It does not come from trying to insulate them from all the failings of other people. Sometimes, it means having awkward conversations (more than once) about difficult subjects, and trusting our kids to learn the right lessons. And that risks failure. Failure of the parents to properly convey the gravity of a situation, failure of the child to properly appreciate the severity of the outcomes. We can’t build an app that makes all the bad things in the world go away, nor should we. All we can do is work with the kids, as parents, as siblings, as relatives, as mentors, and as educators to prepare them for the realities of the world. We do all we can, without relying on heavy-handed legislation or glib explanations of technology which aren’t in the least bit coherent, and we have to trust in ourselves that we did enough to get them through.
