--- Title: The Failure of Account Verification Status: published Date: 2022-11-01 Updated: 2022-11-10 Category: cyber Tags: big tech, platforms, design patterns, twitter ad: elonmusk: @MKBHD Blue check will be the great leveler --- The "blue check" -- a silly colloquialism for an icon that's not actually blue for the at least 50% of users using dark mode -- has become a core aspect of the Twitter experience. It's caught on other places too; YouTube and Twitch have both borrowed elements from it. It seems like it should be simple. It's a binary badge; some users have it and others don't. And the users who have it are designated as... something. In reality it's massively confused. The first problem is that "something": it's fundamentally unclear what the significance of verification *is*. What does it *mean*? What are the criteria for getting it? It's totally opaque who actually makes the decision and what that process looks like. And what does "the algorithm" think about it; what effects does it actually have on your account's discoverability? This mess is due to a number of fundamental issues, but the biggest one is Twitter's overloading the symbol with many conflicting meanings, resulting in a complete failure to convey anything useful. ![xkcd twitter_verification](https://imgs.xkcd.com/comics/twitter_verification.png) # History of twitter verification [Twitter first introduced verification in 2009](https://blog.twitter.com/official/en_us/a/2009/not-playing-ball.html), when baseball man Tony La Russa [sued Twitter for letting someone set up a parody account using his name.](http://www.dmlp.org/sites/citmedialaw.org/files/2009-05-06-La%20Russa%20Complaint.pdf) It was a frivolous lawsuit by a frivolous man who has since decided he's happy [using Twitter to market himself](https://twitter.com/TonyLaRussa), but Twitter used the attention to announce their own approach to combating impersonation on Twitter: Verified accounts. [In their post](https://blog.twitter.com/official/en_us/a/2009/not-playing-ball.html), they announced about a limited beta where Twitter employees would manually verify "public officials, public agencies, famous artists, athletes, and other well known individuals at risk of impersonation." They clarify that, of course, "this doesn't mean accounts without a verification seal are fake" and that "another way to determine authenticity is to check the official web site of the person for a link back to their Twitter account." At this point, the process is clearly focused on impersonation. Verification is to help prevent confusion. But already there's a sneaky conflict buried in the process: it's a beta, and they're starting with individuals who are well known, so already there's some doubt introduced as to the meaning of verification. Does it mean you're who you claim to be, or that you're both that and somehow noteworthy? And who defines *that*? Twitter verification remained a closed don't-call-us-we'll-call-you process [until 2016, when Twitter deigned to let people humbly request that Twitter verify them.](https://blog.twitter.com/official/en_us/a/2016/announcing-an-application-process-for-verified-accounts-0.html) Even with open admissions though, it's still a "[private club, gated behind an invisible, exclusionary admissions process that wasn't documented anywhere.](https://anildash.com/2021/05/20/verifiably-true/)" But even this announcement came with more language about what verification means, which is again problematic: > Verified accounts on Twitter allow people to identify key individuals and organizations on Twitter as authentic, and are denoted by a blue badge icon. An account may be verified if it is determined to be of public interest. Typically this includes accounts maintained by public figures and organizations in music, TV, film, fashion, government, politics, religion, journalism, media, sports, business, and other key interest areas. ... We took a look back and found that the @CDCGov was one of the first Twitter accounts to be verified, in an effort to help citizens find authentic and accurate public health information straight from the source. Ignoring the awkward sentence construction at the top, let's identify the various conflicting goals here: "authentic" again supports this idea of verification as a defence against impersonation, but then "key individuals ... determined to be of public interest" again reinforces this idea of verification being a kind of endorsement given to people an elite committee at Twitter deem to be good and important. They even give "be like the CDC" as an implied example of what it means to be noteworthy. What are people supposed to make of that? Remember the original announcement, where Twitter mentioned that "another way to determine authenticity is to check the official web site of the person for a link back to their Twitter account"? As Anil Dash pointed out in [What it's like being verified on Twitter](https://anildash.com/2013/03/01/what_its_like_being_verified_on_twitter/), Twitter already has a way to verify the authenticity of a Twitter account for analytics: authoritatively linking it to a website. In other words, you could have a badge that confirms "this is the real twitter account of giovanh.com" by just using the existing analytics logic, which would add a layer of trust and identity on top of Twitter's existing username system. (In fact, this is the basic system Mastodon uses.) Why not use a system like this to verify identities once they opened up the process to the public, instead of still requiring a committee to verify people? Because it was already too late to just verify people's identities. After day 1, it wasn't just about impersonation anymore. By this point they had overloaded the "blue check" signifier too much already, and couldn't let "just anybody" be verified. # All the different things the blue check can mean The big problem is that the blue check symbol is wildly overloaded. Verification tries to simultaneously do many different things and therefore fails at all of them because those goals are not aligned with each other. Verification is generally a layer of "trust", but what exactly is that trust supposed to be in? ![MorePerfectUS: On Wednesday, @DannyDevito expressed solidarity with striking Nabisco workers. - “NO CONTRACTS NO SNACKS,” he tweeted. - Today, Twitter stripped him of his verified status, DeVito confirmed to More Perfect Union. https://t.co/rbYfM90sOs](https://twitter.com/MorePerfectUS/status/1428469929288871946) ::: thread unified ![giovan_h: The utter failure of the blue checkmark is Twitter's complete inability to decide what it means and communicate that to people. "This is really who they say they are" is such a home-run concept, but it got loaded with so much baggage that we're... here. https://t.co/6LIREYdX4q](https://twitter.com/giovan_h/status/1428486759940345858) ![giovan_h: If you simplify twitter's verification policy by removing all the contradictory parts, all you get is "It's blue"](https://twitter.com/giovan_h/status/931699586774786048) [Twitter Verification requirements - how to get the blue check](https://help.twitter.com/en/managing-your-account/about-twitter-verified-accounts) reads > The blue Verified badge on Twitter lets people know that an account of public interest is authentic. To receive the blue badge, your account must be authentic, notable, and active. ::: aside It also clearly reads "Twitter does not sell the blue Verification badge. A Twitter employee will never request financial compensation in exchange for a badge or as part of the application process. ... Individuals offering to buy or sell Verification are subject to loss of Verified status and permanent suspension from the platform." which may soon become problematic, thanks to that poorly-placed "never". ## Verification as authenticity That first point is what it's "supposed to" mean: "authentic". That, when you see a user with a name and picture and bio, and you immediately have an idea of who that person is claiming to be, you can look for the checkmark to confirm that you're right. This is already a deeply flawed concept. In order for Twitter to craft a metric that meets that criteria, they have to know who people will think an account is supposed to be, which is of course impossible because that assumption people will make is determined by a function of both the account and the person reading it, not just the account itself. But, just for the sake of argument, let's pretend it's possible to get this right, and not get too down in the weeds of the various bad policies refining *this* is going to produce. But even so, let's take "verification as authenticity" and run with it. ***This super-duper failed.*** Twitter is a company [willing to use their primary verified safety account that alerts users about breaches and major security incidents to do branded content for DC comics.](https://twitter.com/TwitterSafety/status/1484224223555907585) It shouldn't be a huge surprise that they couldn't resist tacking on meanings and connotation to the verified mark either. In a sense, this is a classic linguistic descriptive/proscriptive problem. Does the symbol mean what you say it means, or does it mean what people understand it to mean? This is made worse by the obvious convergent instrumentality at play here. Verification as authentication failed to such a degree that in 2018, they hard-coded "automatically suspend anyone who sets their display name to Elon Musk" into their system to block spam accounts, because the blue check was completely worthless at doing its job. ![giovan_h: elon musk scam account thoughts: - 1. this is the actual problem that twitter verification is supposed to solve - 2. twitter literally just hardcoded "Elon musk" into their system and called it a day](https://twitter.com/giovan_h/status/1023375669487644672) And that's hardly the only case like that. Twitter has a [“government official” flag](https://help.twitter.com/en/rules-and-policies/state-affiliated) that seems entirely orthogonal to its existing verification system, even though ostensibly they're supposed to be doing the same thing: verifying that this account really is "the official voice of the state". That's... that's "verifying" the identity of the account, right? Except verification can't be used for that, because it's already been run into the ground by all the other concepts Twitter couldn't resist tacking onto it, like ## Endorsement ![giovan_h: Twitter: people are confusing verification with endorsement. It's totally not endorsement, you guys - Also Twitter: we added a morals clause that includes your behavior on other websites](https://twitter.com/giovan_h/status/931699237401907200) From that same ["how to get the blue check" page](https://help.twitter.com/en/managing-your-account/about-twitter-verified-accounts), > Verification is currently used to establish authenticity of identities on Twitter. The verified badge helps users discover high-quality sources of information and trust that a legitimate source is authoring the account's Tweets. So it's for "high-quality sources of information". Talk about a load-bearing phrase, they really just drop that and leave. It's no wonder that people feel like the checkmark signals that it conveys credibility to the source, Twitter literally says that it does. What happens when quality and legitimacy come into conflict? That's not a hypothetical. In 2017, Twitter had to pause their verification process entirely after they [verified violent neonazi Jason Kessler's Twitter account](https://www.nytimes.com/2017/11/09/technology/jason-kessler-twitter-verification.html). ::: thread unified ![TwitterSupport: Verification was meant to authenticate identity & voice but it is interpreted as an endorsement or an indicator of importance. We recognize that we have created this confusion and need to resolve it. We have paused all general verifications while we work and will report back soon](https://twitter.com/TwitterSupport/status/928654369771356162) ![TwitterSupport: 2 / Verification has long been perceived as an endorsement. We gave verified accounts visual prominence on the service which deepened this perception. We should have addressed this earlier but did not prioritize the work as we should have.](https://twitter.com/TwitterSupport/status/930926124892168192) At first blush this seems unfair, because verification was supposed to be about authenticity, not an endorsement. And the verified account really was the Twitter account of the real Jason Kessler. But at this point that ship had long sailed already. People were outraged because by this point verification was de-facto endorsement, despite Twitter's claims otherwise. That's why they [refused to verify Julian Assange's account](https://mashable.com/article/twitter-verification-julian-assange). By this point the company understood there was a dual purpose: The little blue tick is both a way to verify people are who they say they are and also convey to users who is a trustworthy source of information. If you're browsing Twitter and you see a tweet from someone you don't recognize, you might not know at first whether you trust the source to be providing reliable information. Then, when you see the tick next to their name, you know that Twitter is endorsing this person as a trustworthy source, and so they're probably telling the truth that [Jews control the media](https://twitter.com/TheMadDimension/status/1586443160296751105). Because Twitter has become a major news platform, this matters *a lot*. To make matters worse, this verification/endorsement happened less than a month after [Jack Dorsey promised more aggressive enforcement against hate groups](https://www.thedailybeast.com/twitter-verifies-charlottesville-rally-boss-jason-kessler-who-called-slain-protesters-death-payback-time). To verify Jason in this environment was utterly tone-deaf, and ultimately representative of much deeper issues within the verification process. As Jack confirmed: ![jack: We should’ve communicated faster on this (yesterday): our agents have been following our verification policy correctly, but we realized some time ago the system is broken and needs to be reconsidered. And we failed by not doing anything about it. Working now to fix faster. https://t.co/wVbfYJntHj](https://twitter.com/jack/status/928658511311097856) So, officially, the internal verification policy still said Jason qualified for verification, even though the public-facing messaging -- even the official messaging! -- said the symbol indicated a high-quality, reliable source of information. ::: thread unified ![anildash: Twitter verifies white supremacists because they wrongly believe verification is an internal corporate process, not a signifier in culture.](https://twitter.com/anildash/status/808427748154712064) ![anildash: This is a classic tech company error — we exist in society, and we don’t get to define our place in society by fiat. The community does.](https://twitter.com/anildash/status/808428184366563328) You'll still frequently find cases like Kessler's, where people complain that Twitter's verification system is being used to platform unconscionable content: ![MichaelEHayden: Fake reporter Jack “Fmr CBS News” Posobiec is once again using the platform @verified gave him to spread lies — this time about Rittenhouse jury: https://t.co/OxRl51YH52](https://twitter.com/MichaelEHayden/status/1460754524084359175) ![im_PULSE: THREAD: The problem with @Twitter and its indiscriminate @verified system, which allows disinformation agents not just to exploit the platform, but to actually thrive on it. This operator linked to the Iranian and Russian propaganda machinery got nearly 40K retweets for this. https://t.co/Iybkb9pCel](https://twitter.com/im_PULSE/status/1507801572377514002) Is this fair? Maybe not, in the imaginary world of 2009, where verification is a moderation tool to prevent impersonation. But in the real world, where verification is a very real and tangible endorsement, it really *is* a problem if Twitter is allowing this endorsement to spread where it shouldn't. It creates responsibility for them, but at this point that's just the check they wrote for themselves. ## It means you're a high-quality, elite person More than just endorsement though, there's a natural conflation between "important" and "good". This is where we start seeing verification used as a status symbol, with verified accounts seen as somehow "elite" as a higher binary tier of users. ::: thread unified ![AstroKatie: The point of Twitter verification is that for certain individuals/organizations it’s useful to be able to verify their statements are coming from them. (This is why so many journalists/reporters are verified.) It’s supposed to help combat disinformation, not be a status symbol.](https://twitter.com/AstroKatie/status/1586892893901520896) ![AstroKatie: People think of it as a status thing because a lot of people with status are verified but the causality is that if you’re well known, you’re more likely to be a target for impersonation and/or there’s more public interest in being able to verify that your statements are yours.](https://twitter.com/AstroKatie/status/1586892897093406720) Especially among less tech-savvy Twitter consumers, the verified badge is seen as a status symbol, rather than the arbitrary Twitter-defined signifier it is. That's why Twitter tried to make it clear they didn't want it to mean status from the outset, because they recognized the danger in correlation. But they didn't do enough to prevent it. Anil Dash discusses in [Verifiably True (2021)](https://anildash.com/2021/05/20/verifiably-true/): > Unsurprisingly, as with anything that's perceived to convey status, verification is one of the parts of Twitter that most consistently inspires resentment, anger, or frustration with the platform. > ... > No surprise, then, that this leads to some measure of conspiracy theories and magical thinking. Regular people outside of the media bubble have developed an entire set of folk beliefs around what verification means, based on what they tell me about why they want to be verified. There are intimations about doing better in Twitter's algorithm, of course — an entirely reasonable assumption. But there's also this broader sense that it opens up a world of possibilities. More than one guy has DMed me saying he needs to be verified because he's about to drop his mixtape, and he wants to make sure people hear it. There's a missing step between a few blue pixels and millions of ears that I don't quite have a grasp on, but I can certainly understand the emotional drive behind it. Twitter is full of little hierarchies: follower counts, likes, retweets, networks. But there's one big stratified social split, and that's verification. It feels like if you can get that badge, you've made it. You're a big shot. ## Power tools and advertisement bonuses, for some reason Having a verified account literally gives Secret Privileges to special users. There are special filters only verified Twitter users have access to (despite these being useful functionality for non-verified users too). These are useful for power-users, but also really valuable tools for companies and advertisers wanting to engage more constructively with the public. And they just don't have access to those features unless they convince Twitter that they're very special boys. ::: aside This used to be even worse -- in the past, verified users had access to additional *trust and safety* features, not just quality of life tools. Verified users also tend to be prioritized (by both Twitter's algorithm and things like external searches) because of the correlation between verification and quality. This adds additional baggage to what "verification means", though, as anything that directly affects visibility is going to be prized for its effects alone, regardless of whatever else it's meant to signify. ## Economic necessity: it means you can pay your bills It's fun to talk about people feeling like being verified will solve their problems, and point out how there's magical thinking around the signifier, but in a way that's not totally untrue. In the internet creator economy, your online social status is, in a very real sense, your livelihood. That's why there's sometimes this intense desperation around verification, why people will be so excited and -- really -- relieved to get it, because it opens very real doors in their professional lives. Again, from [Verifiably True](https://anildash.com/2021/05/20/verifiably-true/): > But there's a broader sense that people's attention, and careers, and opportunities, and even to some degree their lives, are mediated by platforms that are enormously powerful while being fundamentally opaque. Abstractions like "the algorithm" are unknowable, and have millions of people grinding away at a thankless video game that is not only impossible to win, but will actively adapt to keep you from winning if you get too good at it. > > And then, amidst that stress and anxiety and uncertainty, there's a signifier of status. Even better, it's a signifier that's tied to the promise of algorithmic privilege. People will take you more seriously, platforms will amplify your voice, and maybe this entire ecosystem will tilt more fairly in your favor. And that last bit hits at it, I think. People are conscious of how their lives are governed by these giant, opaque, unimaginable systems. Conservatives invent conspiracy theories about censorship because they rightly feel that their lives are unfairly governed by these platforms that are stacked against the user. People are desperate for privilege because of the suffering of being trapped in an economic system that harms those without it. There's a very deep dive into this concept found in [Hearn, A. (2017). Verified: Self-presentation, identity management, and selfhood in the age of big data. Popular Communication, 15(2), 62–77. https://doi.org/10.1080/15405702.2016.1269909](https://www.tandfonline.com/doi/full/10.1080/15405702.2016.1269909), which I'll summarize briefly here: > This new digital “affective” capitalism purloins our desires, emotions, and forms of expressivity and turns them into commodities and assets. Affective capitalism is, quite literally, run on the fuel of individual feeling and self-expression taking place online; self presentation is now a crucial part of the economic infrastructure "Affective capitalism", where capitalism begins intervening in intimate, domestic relationships, creates a framework where creating ways for people to express themselves -- and, in this case, show their legitimacy -- can be turned into an exploitable asset, if you control the systems through which that expression happens. > The verb “verify” generally means > to “confirm,” “support,” or “substantiate” the truth or authenticity of some event, thing, or > person. In rare instances of usage, however, “verify” can also mean “to ‘cause’ to appear > truthful or authentic” (“Verify,” 2002). Following from these definitions, two inflections of the > term “verified” are considered here. The first understands verification as an affirmative > authentication and approval of identity around which users’ desires and affective investments > circulate. The second inflection positions verification as a disciplinary mode of regulation > enacted by a private or state institution that claims authorization over legible and/or “authentic” forms of identity but that, in effect, “causes” legitimate forms of identity to appear. > ... > When users are contacted for verification, they are told they are three quick steps away > from earning their verification badge. The site then takes them through a short quiz > predicated on helping them learn “how to tweet effectively.” The lessons include learning > how to double follower rates by live tweeting events, engaging more followers by asking > them questions and inviting them to a live question-and-answer period, and increasing > likes, retweets, and favorites by including visuals and photos. Finally, Twitter encourages > the user to like and follow other verified accounts in order to increase their own > “truthworthiness.” > ... > The exhortations to learn how to “tweet effectively” in order to receive > verification, however, clearly expose the promotional, self-serving logics of Twitter itself. > In reality, the verification process works to instantiate a new kind of social sorting, or > social class, predicated entirely on a form of “reputation” that Twitter itself defines, > attributes, and then validates in an opaque, unaccountable manner. In this way, Twitter > installs itself as a powerful arbiter of social status and value in a promotional culture and a > “gig economy” where influence and high visibility are increasingly central to job stability > and monetary success. As it comes to challenge more traditional forms of identity authorization, such as passports or medical certificates, however, it must be noted that the checkmark is far from an innocent indicator of a user’s “actual” identity or influence; rather, it is a > careful construction with an entirely instrumental purpose. > ... > The message to those who > seek out and attain the verification checkmark is clear: Build an effective self-brand, cultivate > a following and a reputation, and, most importantly, always be communicating. Here, individuals “are > cast as quasi-automatic relays of a ceaseless information flow”, or > figured as mere data outputs working to ensure “the often serendipitous reunion of > commodities and money”. The grease behind this > constant data generation is provided by free-lunch inducements like the Twitter verification checkmark, whose promise of social status and high visibility encourages users to > perpetually work at posting and crafting themselves online. **Under current volatile economic conditions, however, this kind of attention seeking and identity building is no longer voluntary so much as it is “enforced—a survival discipline for disinvested populations”** ## It means you're paying them money? And all that brings me to the last few hours, where Twitter -- now headed by [known fraud](https://drive.google.com/drive/folders/1zPeWaaCZHqfq0tnkPwc61A6bGHySdj91) [Elon Musk](https://www.thechieftwit.com) -- is looking to reform the verification process by [just charging for it instead.](https://www.theverge.com/2022/10/30/23431931/twitter-paid-verification-elon-musk-blue-monthly-subscription) > The directive is to change Twitter Blue, the company’s optional, $4.99 a month subscription that unlocks additional features, into a more expensive subscription that also verifies users This was originally reported as $20/mo, but Elon Musk looked at rolling that back to $8, after someone on the internet hurt his feelings. ::: thread ![StephenKing: $20 a month to keep my blue check? Fuck that, they should pay me. If that gets instituted, I’m gone like Enron.](https://twitter.com/StephenKing/status/1587042605627490304) ![elonmusk: @StephenKing We need to pay the bills somehow! Twitter cannot rely entirely on advertisers. How about $8?](https://twitter.com/elonmusk/status/1587312517679878144) ![elonmusk: @StephenKing I will explain the rationale in longer form before this is implemented. It is the only way to defeat the bots & trolls.](https://twitter.com/elonmusk/status/1587314744754683905) Based on some disjointed ramblings from Elon Musk, which I guess is what counts as public relations in a full on kleptocracy, the "new vision" for verification looks something like this: ::: thread unified ![elonmusk: Twitter’s current lords & peasants system for who has or doesn’t have a blue checkmark is bullshit. - Power to the people! Blue for $8/month.](https://twitter.com/elonmusk/status/1587498907336118274) ![elonmusk: Price adjusted by country proportionate to purchasing power parity](https://twitter.com/elonmusk/status/1587499283573530625) Of all the even semi-reasonable ways to reform the verification system, this is not one of them. So, Elon sees the existing system as an unfair dichotomy which creates an oppressive class system of elites. His solution to this is to... charge money for a previously free platform feature. So now, instead of verification protecting users from being fooled by impersonation accounts, or telling them what sources of information are reputable, verification would mean you have the disposable income to pay for a social signifier. ::: thread unified ![elonmusk: You will also get: - - Priority in replies, mentions & search, which is essential to defeat spam/scam - - Ability to post long video & audio - - Half as many ads](https://twitter.com/elonmusk/status/1587500060853424129) ![elonmusk: And paywall bypass for publishers willing to work with us](https://twitter.com/elonmusk/status/1587500624098246656) Oh, sorry, it also means you're elevated to a higher social class with a vastly amplified voice in the public square, along with other special powers and privileges compared to the unverified plebeians. This... ends the oppressive class system and solves democracy, somehow? ![elonmusk: @saylor Yes, this will destroy the bots. If a paid Blue account engages in spam/scam, that account will be suspended. - Essentially, this raises the cost of crime on Twitter by several orders of magnitude.](https://twitter.com/elonmusk/status/1587512669359292419) Oh, also, that bit about bots isn't a real concern of his, [that's just a lie he told to try to get out of a bad deal after failing to manipulate the market](https://qz.com/2174898/inside-elon-musks-legal-strategy-for-ditching-his-twitter-deal). But, if this *did* succeed in eliminating bots, it only do so because it required people to pay to effectively use twitter. In other words, it could only do so because it also eliminated vast swaths of legitimate voices that didn't choose to pay Elon for the privilege of having a voice on the internet. ![mmasnick: The whole "priority in replies, mentions" bit is that while he CLAIMS he's giving "power to the people" and removing the classist system, he's actually just making it worse. Those who pay get a cleaned up Twitter. Those who don't get flooded with spam. https://t.co/BioKIHoBF7](https://twitter.com/mmasnick/status/1587575610221555712) **So what happens when you take a signifier you overloaded to mean "good, reputable, high-quality person" and start charging for it?** Well, it definitely loses the original meaning of authentication. It doesn't mean you're really that person, just that you have money. It also loses any signifier of being a high-quality source, if you can just pay for it. It might cause people to pay for the status symbol, or else stop caring about it. It will probably force independent journalists and other reputable, high-quality sources of information out of the designated "reliable" sphere, in favour of low-quality fake news sources that have the funding to pay to amplify their lies and propaganda. It’ll make twitter a more difficult place for everyone to figure out what’s reliable and what isn’t. It's certainly a step in the exact opposite direction of being a free-speech public square, but it's not particularly surprising that Musk was lying about that. # Verification in my remark comments I think the verification design pattern can be a good one. I use it myself, in my blog comments. My account is verified, so people can't impersonate me. That's important. I want to be able to post comments authoritatively, without opening up the door for impersonation. Also, if someone relevant to an article showed up to say something I'd verify them so it's clear they aren't impersonating them. There's definitely value in verification as a design pattern if executed correctly, but how difficult that execution is depends on all the different problems you're trying to use it to solve. (If it's more than one, you've failed right out of the gate.) I think this is another instance where the benefits of scoped rather than universal moderation shine. I don't have to make a decision for the whole internet, just this space. But that's a whole other can of worms. ::: aside update style='--quip: "oh for fuck\'s sake";' *I promise I don't have the gift of prophesy, this stuff is just totally predictable* It's only been a week, but I *have* to come back and talk about how Twitter's attempt at verification reform is going. It's much easier to point at failures and laugh at them than it is to write constructively about things, so I'm going to try to keep this concise. ### November 2 + [Elon Musk is humiliating himself and all we can do is watch in horror - Matthew Sheffield](https://tyt.com/stories/7015be31e708f973a/48c1011463bfab933) + [Hey Elon: Let Me Help You Speed Run The Content Moderation Learning Curve | Techdirt](https://www.techdirt.com/2022/11/02/hey-elon-let-me-help-you-speed-run-the-content-moderation-learning-curve/) ### November 3 As reported in [Elon Musk, Under Financial Pressure, Pushes to Make Money From Twitter - Mike Isaac and Ryan Mac, nytimes](https://www.nytimes.com/2022/11/03/technology/elon-musk-twitter-money-finances.html), Elon Musk's Twitter plans to launch paid verification November 7th, giving all paying subscribers the same "Verified" checkmark as used by verificationh for subscribers but with no kind of identity verification. Twitter [lays off Rumman Chowdhury](https://twitter.com/ruchowdh/status/1588365045158727681), the former director of software engineering who discovered, as part of a long-term research study, that "[Tweets posted by accounts from the political right receive more algorithmic amplification than the political left](https://t.co/ySW6YSx0s8)", a fact that [does not fit into Elon Musk's rhetoric and belief system](https://twitter.com/elonmusk/status/1595114874614390789). Although it seems entirely in-line with what appears to be the man's personal prejudice (as seen in instances such verifying Richard Spencer and Jason Kessler in [Elon Musk's Twitter Verifies Neo-Nazis, QAnoners, Transphobes](https://gizmodo.com/elon-musk-twitter-verify-richard-spencer-jason-kessler-1849773076?utm_medium=sharefromsite&utm_source=_twitter) [as first reported by Emily Gorcenski](https://twitter.com/EmilyGorcenski/status/1590800527872643072)), it actually seems to be a part of the "Deep Cuts Plan", a massive, "delusional" cost-cutting measure reported by [Musk orders Twitter to cut infrastructure costs by $1 billion - Sheila Dang, Paresh Dave and Katie Paul, Reuters](https://www.reuters.com/technology/musk-orders-twitter-cut-infrastructure-costs-by-1-bln-sources-2022-11-03/). ### November 4 Twitter continues and [fires the entire ML Ethics, Transparency, & Accountability team](https://twitter.com/JoanDeitchman/status/1588430085035474944), as well as core members of the [Accessibility team](https://twitter.com/gerardkcohen/status/1588584459779321857). In fact, [Twitter Employees File Class-Action Lawsuit for Mass Layoffs with No Notice - Jules Roscoe, Motherboard](https://www.vice.com/en/article/qjknaq/twitter-employees-file-class-action-lawsuit-for-mass-layoffs-with-no-notice), because Elon pulled the same illegal stunt he tried to pull at Tesla of just firing everybody for an invented cause without paying them their due severance. Back to verification though: Musk [announces in a Tesla-sponsored interview that he plans to algorithmically surpress tweets by non-paying users.](https://www.youtube.com/watch?v=P7wUNMyK3Gs) ::: thread unified ![somebadideas: If you don’t pay the $8 your tweets will be suppressed by an algorithm. Not making this shit up he said it to a room of investors yesterday & claimed this would solve hate speech. “You’ll have to scroll really far to see unverified users” https://t.co/gSzXtzFYtC](https://twitter.com/somebadideas/status/1588876465915166721) ![somebadideas: This wrecks twitter. In his own words caught on camera “you won’t really see’ what your friends are posting - even in your own replies - if they don’t pay for it. - His genius idea is shadowbanning free users?? 😹](https://twitter.com/somebadideas/status/1588939778439139328) ![somebadideas: Here’s them joking about firing half of twitter’s staff yet whining about advertisers claiming ‘we’ve made no change in our operations’. According to ad exec @LouPas who was on a call with him Thurs that is false there’s been massive change to content moderation teams. https://t.co/JqxKUsEGTc](https://twitter.com/somebadideas/status/1588943489114701825) Some good thoughts from [Will Stancil](https://twitter.com/whstancil/status/1590426494190645249) about that, re "the economics of this just literally do not work." ### November 6-7 ![elonmusk: Any name change at all will cause temporary loss of verified checkmark](https://twitter.com/elonmusk/status/1589401231545741312) The "verified badge" shop opened its doors. [Twitter Blue begins advertising a blue check for a monthly fee : NPR](https://www.npr.org/2022/11/05/1134561542/twitter-blue-check-paid-verification-elon-musk) ![elonmusk: Twitter needs to become by far the most accurate source of information about the world. That’s our mission.](https://twitter.com/elonmusk/status/1589413653190938624) Meanwhile Twitter's business isn't looking good either. Read [Musk's Kobayashi Maru - Ed Zitron](https://ez.substack.com/p/musks-kobayashi-maru); revenue is crashing, advertisers have jumped ship, and Twitter is scrambling to hire *back* some of the vital people it illegally fired. Musk is showing himself to be totally unfit to run a business and dangerously so, basically. Astonishingly, as reported by [Elon Musk has discussed putting all of Twitter behind a paywall - The Verge](https://www.theverge.com/2022/11/7/23446262/elon-musk-twitter-paywall-possible), given the projected cut in advertising revenue if Twitter reduces the ads shown to Blue users as it's agreed to, Twitter will actually **lose** money with the new subscription policy the more people sign up. [Two Weeks of Chaos: Inside Elon Musk’s Takeover of Twitter - Kate Conger, Mike Isaac, Ryan Mac and Tiffany Hsu, New York Times](https://www.nytimes.com/2022/11/11/technology/elon-musk-twitter-takeover.html) [‘Economic Picture Ahead Is Dire,’ Elon Musk Tells Twitter Employees - Kate Conger, Ryan Mac and Mike Isaac, New York Times](https://www.nytimes.com/2022/11/10/technology/elon-musk-twitter-employees.html) I'm not just going to repeaet everything reported there, but I do have to point out that apparently in response to high costs, Elon insisted on a payroll audit because he imagined they might be paying "ghost employees", because apparently his only business experience is having once watched office space. In the meantime, Elon himself obsessed over defining his vision of "Free Speech", live on the internet. (He's the only one who gets to talk about free speech though, that's probably fine.) It wasn't good: people [impersonated him as protest](https://www.thedailybeast.com/elon-musk-bans-impersonation-after-parody-elons-flood-twitter), and he didn't handle it well, because he's a small, small man: [The Elon Speedrun Continues; Apparently Comedy Is Not Quite Legal On The New Twitter | Techdirt](https://www.techdirt.com/2022/11/07/elon-musk-continues-to-speedrun-the-content-moderation-learning-curve/) ### November 8 This is where it gets really good. If the "verification" badge doesn't confirm your identity, how are people supposed to safely interact with each other (and, more importantly, brands) on Twitter? At least *someone* at Twitter realized this was a problem, and their solution was... another, secondary badge. ::: thread unified ![esthercrawford: A lot of folks have asked about how you'll be able to distinguish between @TwitterBlue subscribers with blue checkmarks and accounts that are verified as official, which is why we’re introducing the “Official" label to select accounts when we launch. https://t.co/0p2Ae5nWpO](https://twitter.com/esthercrawford/status/1590109344976470016) ![esthercrawford: @TwitterBlue Not all previously verified accounts will get the “Official” label and the label is not available for purchase. Accounts that will receive it include government accounts, commercial companies, business partners, major media outlets, publishers and some public figures.](https://twitter.com/esthercrawford/status/1590109380338647040) [Twitter’s solution for ruining verification is another checkmark - Jay Peters, The Verge](https://www.theverge.com/2022/11/8/23448184/twitter-verification-official-checkmark-gray-blue?utm_campaign=theverge&utm_content=chorus&utm_medium=social&utm_source=twitter) I was surprised to see the "official" field start showing up on my Twitter app though, since I haven't updated it in years. How did Twitter get a new feature out that required a whole new API field so quickly? They didn't. The new "official" tag just wedging in new information to the existing "government figure" field, which I talked about earlier. It's very clear that Twitter has learned *nothing* about the dangers of overloading. ### November 9-11 [Twitter reactivated the new ‘Official’ gray checkmark for accounts that are actually verified - Elizabeth Lopatto, The Verge](https://www.theverge.com/2022/11/10/23452625/twitter-verified-official-blue-gray-check) But for all its flaws, Elon veteoed the second official badge because it wasn't... whatever haphazard idea of own-the-libs he's going for. After this feature briefly launched on November 9, 2022, Elon Musk announced it was canceled and being rolled back. He then stated that the Twitter Blue checkmark would be the only "verification" badge. ::: thread ![elonmusk: @MKBHD I just killed it](https://twitter.com/elonmusk/status/1590383366213611522) ![elonmusk: @MKBHD Blue check will be the great leveler](https://twitter.com/elonmusk/status/1590383539610324992) Elon Musk says Twitter scrapped the "Official" tag literally hours after its launch. But then Esther has to come back and publicly debunk him, because he's just a dangerous bean bag. ![esthercrawford: @nelson The official label is still going out as part of the @TwitterBlue launch -- we are just focusing on government and commercial entities to begin with. What you saw him mention was the fact that we're not focusing on giving individuals the "Official" label right now.](https://twitter.com/esthercrawford/status/1590388608359628800) ::: thread ![elonmusk: Please note that Twitter will do lots of dumb things in coming months. - We will keep what works & change what doesn’t.](https://twitter.com/elonmusk/status/1590384919829962752) ![esthercrawford: There are no sacred cows in product at Twitter anymore. Elon is willing to try lots of things -- many will fail, some will succeed. The goal is to find the right mix of successful changes to ensure the long-term health and growth of the business. https://t.co/cMf27EmmpJ](https://twitter.com/esthercrawford/status/1590386711179464705) ![elonmusk: As Twitter pursues the goal of elevating citizen journalism, media elite will try everything to stop that from happening](https://twitter.com/elonmusk/status/1591121142961799168) He's just throwing stuff at the wall, except the wall is Twitter's production environment. It's astonishing. ::: thread unified ![giovan_h: Incredible to see a CEO pushing his policy whims onto Twitter within a few hours without any sort of internal policy process. Can you imagine what Trump would have given for that kind of power? https://t.co/HuUeFYTVPq](https://twitter.com/giovan_h/status/1589838772182474752) ![giovan_h: Major, major changes to core functionality without even a news announcement on a company blog somewhere. Astonishing behavior for a company like this.](https://twitter.com/giovan_h/status/1589839163020525568) Twitter's business side is still flailing, of course. [Twitter lawyer warns that Elon Musk is putting company at risk of billions in FTC fines - Alex Heath, The Verge](https://www.theverge.com/2022/11/10/23451198/twitter-ftc-elon-musk-lawyer-changes-fine-warning) features the absolutely *flawless* line > The Verge reached out to Musk for comment. Twitter no longer has a communications department. [Musk seeks to reassure advertisers, promises rapid changes to Twitter - Gerrit De Vynck, Jacob Bogage and Faiz Siddiqui, The Washington Post](https://www.washingtonpost.com/technology/2022/11/09/elon-musk-twitter-advertisers/) [Twitter’s content moderation head quits as departures alarm the FTC - Joseph Menn, Cat Zakrzewski, Faiz Siddiqui, Nitasha Tiku and Drew Harwell, The Washington Post](https://www.washingtonpost.com/technology/2022/11/10/twitter-security-resignations/) ::: thread unified ![KurtWagner8: NEW: Hearing that the top execs remaining at Twitter -- head of Trust & Safety @yoyoel and head of sales @robinw -- both resigned today - When you add in that Musk told employees this afternoon that bankruptcy is possible, it feels like Twitter is almost collapsing in real-time](https://twitter.com/KurtWagner8/status/1590824366492643328) ![KurtWagner8: There were the two execs that joined Elon on Spaces yesterday when they tried to soothe advertiser concerns. Hearing from a number of folks asking: "Who is left?" - Hard to overstate how gutted Twitter's leadership ranks have become](https://twitter.com/KurtWagner8/status/1590828164443701248) ![KurtWagner8: Plot twist! I’m told while Wheeler did indeed resign earlier today, Elon convinced her to stay. https://t.co/GKPnyi0DZx](https://twitter.com/KurtWagner8/status/1590895999362662402) It's so obviously bad that Elon Musk's personal lawyer, who has previously stated on the record that "Elon Musk sends rockets into space, he doesn't need to worry about the FCC") had to mass-email the remaining employees to say they won't go to jail if they follow Musk's commands. That's really, really not an email you want to get. [Musk’s Lawyer Calms Twitter Staff Fearing Jail Risk for FTC Lapses - Ryan Gallagher and Kurt Wagner, Bloomberg](https://www.bloomberg.com/news/articles/2022-11-11/musk-lawyer-calms-twitter-staff-fearing-jail-risk-for-ftc-lapses?sref=IJ2uXcA2&leadSource=uverify%20wall) Elon's response is to openly antagonize his oversight. He's not a smart man. ::: thread ![SenMarkey: A @washingtonpost reporter was able to create a verified account impersonating me—I’m asking for answers from @elonmusk who is putting profits over people and his debt over stopping disinformation. Twitter must explain how this happened and how to prevent it from happening again. https://t.co/R4r7p6mduP](https://twitter.com/SenMarkey/status/1591164312168824832) ![elonmusk: @SenMarkey @washingtonpost Perhaps it is because your real account sounds like a parody?](https://twitter.com/elonmusk/status/1591813228119855104) ![SenMarkey: One of your companies is under an FTC consent decree. Auto safety watchdog NHTSA is investigating another for killing people. And you’re spending your time picking fights online. Fix your companies. Or Congress will. https://t.co/lE178gPRoM](https://twitter.com/SenMarkey/status/1591827463583453190) --- So remember how the entire purpose of verification was to combat impersonation on Twitter? How do you think that's going, now that you can buy a badge that says "Truth" with money instead of credibility? Really, really bad. ![This fake twitter account is impersonating twitter corporate with a newly purchased check mark](https://pbs.twimg.com/media/FhIwGfMXoAAdi2n.jpg) [Twitter's New Paid Verified Check Causes Frenzy After Fake LeBron Trade Tweet (www.tmz.com)](https://www.tmz.com/2022/11/09/twitters-paid-verified-check-fake-lebron-james-trade-tweet/?adid=social-twa) ![Rudy Giulianni](https://pbs.twimg.com/media/FhKEBvsUAAA_Fp-.png) ![DiscordianKitty: I want to be at Nintendo's board meeting, a company that cares deeply about brand safety and its reputation for family values, as they're shown this tweet of the owner of the company currently not protecting their brand gloating that he's getting paid for it. https://t.co/mz2vz8QqNk](https://twitter.com/DiscordianKitty/status/1590603831749931008) ![imraansiddiqi: This checkmark thing is going well. https://t.co/6MVlAaqI2O](https://twitter.com/imraansiddiqi/status/1590604863112904704) ![mkeallison: @wongmjane curious why https://t.co/xAfgTC63pR](https://twitter.com/mkeallison/status/1590961921209294849) ![MisterRatt: Twitter Blue is going about as well as everyone predicted, and it's an amazing spectacle to watch. Like a train crash filled with glitter. https://t.co/Icb3vru3Ca](https://twitter.com/MisterRatt/status/1590632287544438784) ![drewharwell: Another fake-but-verified Twitter account. This time, it's one of the world's biggest military contractors. What could go wrong? https://t.co/MmbpcAQa1H](https://twitter.com/drewharwell/status/1590890265879154688) ::: thread unified ![drewharwell: Spokesperson for pharmaceutical giant Eli Lilly says they're in communication with Twitter to address the fake-but-verified Eli Lilly tweet that has been up for three hours and has 1,500 retweets and 10,000 likes https://t.co/Ai3lq50YRh https://t.co/zIWDsd8c0B](https://twitter.com/drewharwell/status/1590822878274097152) ![drewharwell: @elonmusk The fake Eli Lilly free-insulin tweet has now been online for six hours. 3,000 retweets. - The (actual) company responded three hours ago. But, hey, at least there's a "Community Note" now. - And all this for a crisp $8. Enjoy it, @elonmusk! https://t.co/8XiIz4EH5L](https://twitter.com/drewharwell/status/1590870708824920064) ![LillyPad: We apologize to those who have been served a misleading message from a fake Lilly account. Our official Twitter account is @LillyPad.](https://twitter.com/LillyPad/status/1590813806275469333) [Eli Lilly\['s stock price\] Dives After Fake Twitter Account Promises Free Insulin; Takes Novo Nordisk, Sanofi With It - Allison Gatlin, Investor's Business Daily](https://www.investors.com/news/technology/lly-stock-dives-taking-novo-sanofi-with-it-after-fake-twitter-account-promises-free-insulin/) ![elonmusk: Twitter needs to become by far the most accurate source of information about the world. That’s our mission.](https://twitter.com/elonmusk/status/1589413653190938624) It's a specatcular train wreck. We've left "tricky design problem" far behind, we're well into "literal farce". And it was was all completely, easily avoidable. Elon Musk is just a damn fool. ::: thread unified ![whstancil: The thing about paid verification is that it doesn’t provide ANY legitimate utility. There’s no non-sordid use case. - Its MAIN USE is social trickery, either by artificially increasing the stature of an account, or worse still, enabling outright impersonation. https://t.co/UMRpNrgcOo](https://twitter.com/whstancil/status/1590742560393531392) ![whstancil: Elon and his clique of idiot culture warriors appear to have completely missed this because they were/are incapable of seeing the blue checks outside the extremely narrow lens of social resentment against a smarter, poorer media power class.](https://twitter.com/whstancil/status/1590742940053561344) Internally, Twitter distinguishes between real verification and vanity badges, so people have written browser extensions to let people browse safely since Twitter has decided to endanger them. + [GitHub - wseagar/eight-dollars: A browser extension that shows twitter blue vs real verified users](https://github.com/wseagar/eight-dollars) + [GitHub - wesbos/who-blue: A browser extension to separate lords from peasants](https://github.com/wesbos/who-blue) And rat boy is *still* out there writing up ideas that someone stoned out of their gourd might dream up in the shower and quickly discard and releasing them as official policy statements for the biggest social media company in the world. ::: thread ![elonmusk: @TomFitton @Twitter @yoyoel @JudicialWatch Twitter will not censor accurate information about anything](https://twitter.com/elonmusk/status/1588666492572553216) ![giovan_h: it's really just such low-hanging fruit at this point. "How does this policy interact with your other policies?" "Does this mean Twitter isn't removing leaked information or doxxing?" but it's unfair to ask those because of course he can't answer those questions, he's a bean bag](https://twitter.com/giovan_h/status/1590869008936755200) ::: thread ![elonmusk: Going forward, accounts engaged in parody must include “parody” in their name, not just in bio](https://twitter.com/elonmusk/status/1590884973535711232) ![elonmusk: To be more precise, accounts doing parody impersonations. Basically, tricking people is not ok.](https://twitter.com/elonmusk/status/1590886170543915009) ![Enichan: I'm going to give everyone access to a checkmark! That'll show those mean journalists! - *has to make 500 new arbitrary rules all with their own edge cases that can be exploited in turn which can't be enforced because I sacked 98% of the moderation teams*](https://twitter.com/Enichan/status/1590891474094555136) ![dril: can we all please affix "#justforfun" to the tweets that aren';t meant to be taken seriously. this would really cut down on the mishmash](https://twitter.com/dril/status/12932820075) As of 11/11, it looks like Twitter is suspending the launch of Twitter Blue entirely, and is now actively trying to prevent new subscriptions. ::: thread unified ![ZoeSchiffer: NEW: Twitter has suspended the launch of Twitter Blue and is actively trying to stop people from subscribing "to help address impersonation issues," per an internal note. 1/](https://twitter.com/ZoeSchiffer/status/1591081913166745601) ![ZoeSchiffer: The announcement was posted on Slack: "An update on what we did tonight: hid the entry point to Twitter Blue, added the 'official' label for ONLY advertisers. Note: here is at least one way for users to sign up for Blue. Legacy Blue users can go to subscriptions and upgrade" 2/](https://twitter.com/ZoeSchiffer/status/1591082807237185539) ![elonmusk: Holding off relaunch of Blue Verified until there is high confidence of stopping impersonation. - Will probably use different color check for organizations than individuals.](https://twitter.com/elonmusk/status/1594861031670820864) [Musk: Paid checkmarks won’t return until Twitter can stop impersonation | Ars Technica](https://arstechnica.com/tech-policy/2022/11/musk-paid-checkmarks-wont-return-until-twitter-can-stop-impersonation/) # Related Reading - [Nitasha Tiku, "Twitter's Authentication Policy Is a Verified Mess", 2017](https://www.wired.com/story/twitters-authentication-policy-is-a-verified-mess/){: .related-reading} - [Content Moderation Case Study: Twitter Removes 'Verified' Badge In Response To Policy Violations (2017) | Techdirt](https://www.techdirt.com/2021/09/15/content-moderation-case-study-twitter-removes-verified-badge-response-to-policy-violations-2017/){: .related-reading} - [Dan Phiffer, "How We Verified Ourselves on Mastodon — and How You Can Too"](https://themarkup.org/levelup/2022/12/22/how-we-verified-ourselves-on-mastodon-and-how-you-can-too){: .related-reading} - [Geoffrey A. Fowler, "Twitter said it fixed ‘verification.’ So I impersonated a senator (again)"](https://www.washingtonpost.com/technology/2023/01/05/twitter-blue-verification/){: .related-reading} - [Sara Fischer Rebecca Falconer, "Verified" becomes a badge of dishonor (2023)](https://www.axios.com/2023/04/23/verified-checkmark-twitter-badge){: .related-reading} - [Matt Blinder, "Dril and other Twitter power users begin campaign to 'Block the Blue' paid checkmarks" (April 2023)](https://mashable.com/article/block-the-blue-twitter-campaign-dril){: .related-reading} *fine, I'll separate out the Elon Musk stuff* - [Matthew Sheffield, "Elon Musk is humiliating himself and all we can do is watch in horror"](https://tyt.com/stories/7015be31e708f973a/48c1011463bfab933){: .related-reading} - [Mike Masnick, "Hey Elon: Let Me Help You Speed Run The Content Moderation Learning Curve"](https://www.techdirt.com/2022/11/02/hey-elon-let-me-help-you-speed-run-the-content-moderation-learning-curve/){: .related-reading} - [Paris Marx, "Elon Musk's Flawed Vision and the Dangers of Trusting Billionaires"](https://time.com/6203815/elon-musk-flaws-billionaire-visions/){: .related-reading} - [Elon Musk, Under Financial Pressure, Pushes to Make Money From Twitter](https://www.nytimes.com/2022/11/03/technology/elon-musk-twitter-money-finances.html){: .related-reading} - [Nilay Patel, "Welcome to Hell, Elon"](https://www.theverge.com/2022/10/28/23428132/elon-musk-twitter-acquisition-problems-speech-moderation){: .related-reading} - [John Bull on the "Trust Thermocline" (thread)](https://twitter.com/garius/status/1588115310124539904){: .related-reading} - [Bak-Coleman, J. B. et al, E. U. (2021). Stewardship of global collective behavior. Proceedings of the National Academy of Sciences, 118(27).](https://www.pnas.org/doi/10.1073/pnas.2025764118){: .related-reading} - [Justin Jackson, "Is Twitter Done?"](https://justinjackson.ca/is-twitter-done){: .related-reading} - [Why Is This Happening? The Chris Hayes Podcast, "Twitter's Elon Musk Era with Kara Swisher"](pdst.fm/e/nbcnews.simplecastaudio.com/59eb82e8-198b-4b11-b64a-c04a9083812d/episodes/e63b6b15-7b19-45c7-b83e-2e30d755609d/audio/128/default.mp3){: .related-reading} - [Ed Zitron, "The Fradulent King"](https://ez.substack.com/p/the-fraudulent-king){: .related-reading} - [Christia Peterson, "twitter blue screenshot storyline"](https://twitter.com/christapeterso/status/1592317592966168576){: .related-reading} ![ZenOfDesign: The ironic thing is that charging for verification would be VERY good for Twitter. If Twitter charged a ONE-TIME fee of $20 bucks, and spent that money actually verifying that people were who they said they were, a ton of people would likely sign up for that. - 12/](https://twitter.com/ZenOfDesign/status/1649524935331569664)