Bluesky’s launch of a verification system has raised a slew of questions among its user base from who will be picked — and why — to what outside organizations might be involved and whether the self-verification process will end? TechCrunch has some answers.
After a leak last week, Bluesky officially announced Monday the arrival of its verification system, which is designed to ensure that notable people are who they say they are on the social network. While similar in some respects to the system that Twitter had once used before pivoting to paid verification under Elon Musk, Bluesky’s verification builds on the company’s decentralized ethos by giving other entities the authority to verify users independently.
Bluesky briefly explained how this process worked in a blog post. However, the numerous questions found in the replies to Bluesky’s posts and on other forums, like Reddit, indicate that many don’t fully understand verification.
We’ve rounded up some of the more common questions that seem to be on users’ minds to try to help explain the verification process and its rollout.
While Bluesky itself didn’t respond to TechCrunch’s requests for comment, we’re referencing the company’s own documentation and its leaders’ posts to try to answer the questions we found many were asking.
Which organizations have been given the ability to verify besides Bluesky itself?
One missing piece of information from Bluesky’s blog post was which organizations outside of Bluesky are being given the tools to verify others. The company explains that “Trusted Verifiers” will be those organizations that Bluesky allows to issue blue checks. However, the only example of this that was shown was The New York Times, which is now allowed to issue checks to its own journalists.
What Bluesky hasn’t said is what other organizations now have this power, or how they were chosen.
It’s also unclear if Bluesky has others lined up for early access; the company didn’t reveal any future partners on this effort. This makes Bluesky’s announcement of verification feel somewhat premature, as people want to see the working examples of decentralized verification systems from the get-go, even if it’s initially with only a few testers.

We’ve found that beyond The New York Times, there are only a couple of other organizations that have been given Trusted Verifier status at this time. These include Wired and The Athletic, according to Bluesky software engineer Samuel Newman.
Bluesky itself said it’s only starting out with a “small and non-comprehensive group of accounts.”
CEO Jay Graber added that the initial set includes “news orgs” that have agreed to verify their journalists. Later, the company will launch a form that will allow other organizations to apply for Trusted Verifier status, but Bluesky has not indicated when this will happen.
Does Bluesky want verification to be perceived as an indicator that a person is “trustworthy” too?

One of the problems with Twitter’s old verification system is the verification badge became a coveted achievement. If a user was noteworthy enough, had gained a certain number of followers, or was an expert in their field, they began to feel they should be verified. But they may not have received a verification badge because Twitter disagreed with them about their importance. Other times, they just didn’t know the right people at Twitter to ask.
How Bluesky feels about this matter is still somewhat vague. The company’s blog post suggests that its own verification will work to “proactively verify authentic and notable accounts,” but doesn’t explain the criteria it’s using to deem an account “notable” enough to be worthy of verification.
When it was suggested to Bluesky CTO Paul Frazee that people were curious about the who, when, and why around who Bluesky was choosing to verify, he only responded by saying “yep yep.” How mysterious!
We’ll mark this one “TBD.”
Why is [X] verified but not [Y]?
As with any launch of verification, people are quick to focus on the haves and the have-nots. Who got verified first and why? And why did this person or organization get verified over that one?

When Bluesky announced that verification had launched, many assumed they would immediately see blue checks everywhere. Instead, people were struggling to understand how news organizations like CNN, The WSJ, and Bloomberg received the blue-and-white verification badge but others like Politico or MSNBC did not (as of the time of writing!).
This is likely because Bluesky has not completed its rollout of verification.
The company suggested in a post on the network that the launch of verification was not an instant process, when it announced that users would “start seeing” blue checks appear in Bluesky following the launch.
Plus, Bluesky noted it wasn’t accepting any verification requests “currently,” which could imply that, at a later date, it will have a process for doing so.
In other words, we wouldn’t read too much into who is verified as of today or what that means, as we’re only at the beginning of this rollout.
What does this mean for self-verification?
Before the launch of verification, Bluesky offered another way for users to verify that they are who they say they are: with domains. In 2023, the company began allowing organizations and individuals to set a domain as their username, and since then, over 270,000 accounts have done so, Bluesky says.

This system is not going away, Bluesky notes, as it will “continue to be an important part of verification on Bluesky.”
Instead, it will now become another layer of verification. It will remain an optional way to confirm an identity, however, not a mandate for getting verified.
Still, Bluesky says it “highly recommends” official organizations and high-profile individuals do this and provides how-to documentation to get started.
Also worth noting: the company said in December 2024 that after someone changes a Bluesky username to a website URL, their old bsky.social username will still be reserved for them. This prevents account impersonation by bad actors. For that reason, the reservation will never expire.
What does this mean for those accounts that were unofficially verifying users ahead of the official launch of verification?
As one example, Hunter Walker and Guan Yang have been running a labeler to unofficially verify a range of diverse media outlets, large and small, national and local. This labeler also verified elected officials, prominent activists, political operatives, and other celebrities, according to its website.

Walker noted in a post on Bluesky the company had not approached him about the official verification system.
“No one at Bluesky has ever talked to me about verification. I sure have a lot of legwork done if they’d like to!,” he wrote.
So far, these “unofficial” labels and badges have not disappeared from users’ accounts, but the future of the labeler remains uncertain.
“We will keep going if it’s needed but, frankly, I hope this ultimately moves to a single, clear source of credible verification,” Hunter said on Bluesky.
Rapper and musician Flavor Flav had also helped to confirm which Bluesky accounts were authentic in the social network’s earlier days. His guidance may also no longer be required once the verification system fully rolls out. (Unless he becomes a Trusted Verifier on his own, we suppose!)
What will happen if a Trusted Verifier begins abusing its privilege?
In theory, those granted a Trusted Verifier status should be … trustworthy. But what happens if one goes rogue and begins to operate unethically, like taking payments for verification?
There’s no detailed process for how this will be handled, nor is there an official set of rules that Trusted Verifiers have to agree to as of yet — as far as we can tell.
But it appears that Bluesky has considered the possibility.
When asked in a Bluesky post what would happen in the case that a Trusted Verifier abused its privilege in some way, CTO Paul Frazee responded that “Bluesky can intervene if needed.”
Why is the symbol a blue-and-white check, like Twitter, instead of a blue Butterfly or something unique?
Lastly, some people wondered if adopting a similar look-and-feel to Twitter’s checks made sense. Bluesky is different, so shouldn’t it use some other design language?
Frazee replied to questions about this, too, saying at different times the company had tried other colors, like green, but found that blue better matched Bluesky’s existing color palette.
The team has not officially explained why it went with a traditional checkmark instead of a blue butterfly, for instance, as many have suggested. Likely, though, it’s just a matter of the check being a recognized symbol with an understood meaning.