After pausing the release to avoid disrupting the U.S. midterm elections, Twitter launched their new verification system on Wednesday: Anyone willing to pay 8 bucks a month for Twitter Blue would now receive the long coveted (by some) “blue check mark,” which, of course, is actually a white check mark against a blue background.
The rollout has been an unmitigated disaster.
People began signing up for the service immediately, yes, but went on to impersonate celebrities, politicians, and corporations. People like Donald Trump and Lebron James, companies like Nintendo and Twitter itself.
Today, day two of the rollout, the fiasco continued.
Early this afternoon, for example, a rogue account—looking pretty authentic with the handle @EliLillyandCo—tweeted a short message with huge impact: “We are excited to announce insulin is free now.”
At 4:45pm, WaPo’s Drew Harwell was reporting that Eli Lilly (the actual company) was “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.”
The thing is, of course, this new verification system isn’t a verification system. In fact, Twitter planned on displaying a second gray badge to signal actual authenticity but Musk killed the feature just hours after it launched. He was putting all his chips on the blue badge, saying it would be “the great leveler.”
In fact, from a design perspective, he simply created a usability nightmare.
Let’s look at the user experience.
Technically, you can still check to see if a Twitter account has been truly verified by clicking on the blue badge, which launches an overlay declaring, “This account is verified because it’s notable in government, news, entertainment, or another designated category.”
On the other hand, if you click on the badge on an account, which hasn’t really been verified, you see a message which reads “This account is verified because it’s subscribed to Twitter Blue.”
Sorry, but that’s not verification. That’s buying a 30×30 pixel badge.
So Twitter now presents you with two wildly different reasons for the same badge, explained behind a click you have to know exists to activate. And by the way? This information can’t be accessed on individual tweets. You have to go to the account’s profile to activate it. If you even know to activate it.
This pattern utilizes what’s referred to as “progressive disclosure” to reveal information if you want to see it, but to remove clutter from an interface if you don’t need to see it. It’s often a good pattern. Except when you’re placing important information a click away, information such as “Is this a real person or company?”
This is elementary UX design stuff.
Elon Musk can’t say he wasn’t warned. Critics of the feature’s new direction pointed out well before it launched that the new policy meant anyone could pony up $8 and change their name to whatever they wanted and it’d be displayed with the verification badge placed immediately after it, potentially causing mass confusion, since for years, the badge has meant that the account you’re looking at was, at least, authentic. That it had been verified.
Overnight, that was no longer the case.
More broadly? This is what happens when design decisions are made without concern for the moral and ethical implications of the resulting changes. Musk has essentially rolled out the social media equivalent of a “dark pattern.”* On purpose.
In response to some very loud voices, who had been clamoring for access to the blue badge for years, Musk rushed this feature to the market. Now, it offers a case study in what happens when you allow people to purchase authority, instead of earning it.
As a result, NPR is now advising their reporters not to give out their Twitter handles on air but not to delete their accounts because they could be taken over and impersonated with a blue badge.
Others point out the change poses national security risks as people have begun posing as international figures.
One data scientist noted that “a variety of Twitter accounts with GAN-generated faces have paid $8 for a blue checkmark.” GAN stands for “generative adversarial network” and it’s the type of AI/machine learning framework used by sites such like thispersondoesnotexist.com. These are the types of accounts, of course, that many are concerned will share disinformation—which will now accompanied by the blue check mark.
All of this is happening in the wake of Musk laying off 3,700 Twitter employees. And last night saw the exodus of Twitter’s chief information security officer, chief privacy office, and chief compliance officer. Twitter no longer has a communications department. Musk also fired the company’s entire board leaving himself as its sole member. He fired so many people, Twitter reportedly had to reach out to dozens of employees in an attempt to hire them back. Turns out they needed the help. The company is in disarray.
It’s like watching a city burn down to the ground.