Bluesky 2025 Transparency Report
bsky.social
bsky.social
https://bsky.social/about/blog/03-12-2024-stackable-moderati...
Now in 2026, their transparency report says nothing about stackable moderation or moderation services. I guess nobody is using them, at least not in significant enough volumes that it would have a meaningful effect on the at least not enough for them to have a meaningful effect on the Trust & Safety team's job.
Likewise, they tout "thousands of Personal Data Servers operated by people across the federated AT Protocol network", but that's out of "41.41M users".
It's fine, I guess. It's just not meaningfully decentralized.
It’s spread out over subgroups and niches. I imagine the biggest independent moderation service is blacksky’s, and they’re not exactly best friends with bluesky.
I use about five different moderation services, and a handful of independent blocklists.
> It's fine, I guess. It's just not meaningfully decentralized.
It’s better than the situation on X.
i find it’s pretty toxic, in a militant way… that doesn’t get moderated of course.
That’s “free expression” when it’s about topics blueskyers all agree on.
waits to be called a nazi
Which is fine -- build what you want to be a part of -- but don't pretend it's the neutral position.
And no, I'm not talking about "1939 Germany" views.
"The deportations will continue regardless of the tantrums of the hysterical and mentally ill."
Which ironically is similar to some "1939 Germany" views.
That comparison is precisely the problem: it distorts history, inflates moral claims, and shuts down serious discussion.
This is also largely the standard level of rhetoric on Bluesky, which is fine -- but manufactured consensus on a heavily moderated platform is not the same thing as factual or moral authority.
Fortunately the Republicans, specifically Tom Homan and Stephen Miller, can be shown to have a record of dehumanizing people to the point of cruelty.
As far as anyone could tell they were behind the effort to separate children from their families, and the decision to intentionally destroy records, and prevent the recording of data, which ultimately left over a thousand children orphaned to this very day.
I know someone in CBP who volunteered to try help sort that situation out, ready to get on a plane, paying out of pocket, and they were told to stand down by leadership.
The republican-led executive branch wanted to inflict pain despite the law, and despite "policy".
And now those same people (Homan and Miller) are behind the door-to-door raids, asking people for their papers, building detention centers(even though we're supposed to be sending them back home...), and targeting political enemies.
Obama managed to deport 3 million people without this excess use of cruelty, civil rights violations, manpower, or money.
This is something else.
This level of hatred towards the other is the type of seed that may or may not grow into a holocaust. It's understandable if some people want to kill it before it sprouts by drawing obvious parallels.
Holocaust analogies based on unsupported anecdote and asserted intent aren’t analysis; they’re unfalsifiable rhetorical escalations designed to end debate. If every disliked policy is treated as a "seed" of genocide, as is now common, the term loses meaning and becomes an empty rhetorical weapon. Argue specific actions with evidence and standards, or don’t -- but stop inflating unfalsifiable moral claims to the point where serious critique is impossible.
> Even if the current state of immigration policy was forged in a bipartisan agreement(it wasn't), it would be inhumane and I would condemn it.
We haven’t passed a comprehensive immigration law since 1986, and the enforcement framework in use today arises from bipartisan legislation passed in 1986, major subsequent revisions in 1996, and layers of later executive discretion exercised by administrations of both parties. We had four years of functionally non-existent enforcement, and while I cannot ascribe motive, the natural outcome was to make later enforcement incredibly difficult -- a consequence that is now plainly visible.
If you think those laws are unjust, argue that -- but don’t pretend this is some novel or uniquely partisan creation.
There is no need to pretend. The specific excesses I mentioned are Republican actions. The Democrat failures do not excuse the illegal and immoral Republican cruelty.
Help me out here. Its not clear to me what the Left considers a failure.
My own state promptly made it illegal for local law enforcement to cooperate with border patrol or immigration enforcement agents in any circumstance.
So now, if we want our country to have meaningful borders, immigration enforcement has to be done the hard way, and it shouldn't be surprising that Kristi Noem's clown show is showing signs of clusterfuckery. It's actually surprising that things have gone as well as they have.
every dem president has to clean up after a republican president
And of course, not to be outdone by Obama, Joe Biden managed to lose control of the border in a manner that has no precedent in American history, once again handing Trump an easy victory. It's like they wanted to lose another election.
To grasp how radically the Democrat party has moved left on immigration, recall that Obama and Hillary Clinton ran against each other on strict immigration enforcement. Here's my favorite Hillary quote from a 2008 campaign speech:
> "If they’ve committed a crime, deport them, no questions asked. They’re gone. If they’re working and law-abiding, we should say here are the conditions for you staying: You have to pay a stiff fine because you came here illegally, you have to pay back taxes, you have to try to learn English, and you have to wait in line."
Difficulty does not equal unconstitutionality. If you think the measures required violate specific constitutional provisions or civil-rights protections, name them and explain how.
Otherwise, this is an empty moral reframe that clumsily sidesteps concrete claims about enforcement feasibility and changed conditions.
You can't complain about content on bluesky because unlike every other platform you must choose which feeds you use.
Pretty shitty that ones choice of social media is so politicized but if you must pick a side... I will pick the non-nazi side, thank you very much!
I’m growing tired of those bans on legal content that isn’t inherently harmful (we are talking about fictional humanized animals here) but considered “icky” by platforms and payment processors.
So I don’t care if the AT protocol is technologically superior to ActivityPub (?) – the Mastodon community has a healthier moderation and mindset than Bluesky, in my opinion.
Furry art, including quite explicit furry art, is very common on bluesky and doesn’t seem be especially restricted by policy. I mean, unless they also happen to be depicting nonconsensual sexual interactions, an orthogonal concern to the furry aspect.
> I’m growing tired of those bans on legal content that isn’t inherently harmful (we are talking about fictional humanized animals here) but considered “icky” by platforms and payment processors.
Well, you are free to avail yourself of the forums that lack those policies. Now, I know you’ve complained that they are “harmful”, but... Maybe there is a reason that other forums choose to put bans in place.
Now there's an understatement. It's bloody impossible to get rid of. People here are sneering at all the political content but they're ignoring the curvaceous elephant in the room. I think maybe bsky has improved things now, but a while back their adult content filters were not up to the task. When I first made an account I almost gave up on it because until I got all the right filter words set up it was nothing but weird porn whac-a-mole (actually that's probably a poor choice of words...)
Small aside, consent also depends on the jurisdiction – in mine, it must be verbal, so it means that if I were to draw a situation which involves a character being forced to do something but showing their consent non-verbally, it would still be non-consensual, and thus, forbidden by Bluesky’s terms of service if the PDS was hosted in my jurisdiction.
Anyway, my point is that all those illustrations should be properly labeled, but not necessarily forbidden by Bluesky’s ToS. As I understand it, fictional non-con content being banned by Bluesky means that even hosting it on one’s PDS is a no-go.
so it's _not_ that hard.
But Twitter felt cringe to me long before it was consumed by Musk and politics. Messing with the feed has backfired all of the big platforms. First Facebook then Twitter and most recently Instagram.
They all became a closed loop of content that is force fed. Injecting an ad in the a feed we control wasn't ever enough.
But I remember the early days of MySpace and Facebook with a certain nostalgia, and I'm pained to see the current state of all these tools. Such a thorough report as this gives me a little hope that perhaps an acceptable middle ground can be found for Internet communities at large scales.
I don't think I'll be hopping back in any time soon, but perhaps the research and positive advancements in social media aren't over yet.
It would be nice to see some more transparency around the decisions of whether an account gets verified or not. So far it’s feeling like a “cool club” with little rhyme or reason with regards to certain verification decisions.
Related to that is also the need to add more trusted verifiers. Are there any plans to allow third parties to provide verification services or is it always only going to be journalistic and educational institutions?
That's how Twitter started, and when the policy changed, the "cool club" members threw public tantrums (some of which still seethe to this day).
It's all very high-school cafeteria clique to me.
Anyone can put together a moderation service that labels accounts that they’ve vetted or blacklisted. It wouldn’t be that taxing to host one, but the labor to maintain it is a different story.
Worth keeping in mind, the original plan was to entirely rely on domain names being used as handles for verification. All ~500 .gov handled accounts on there are almost certainly who they claim to be, unless they are able to set DNS records on those domains.
I have a look at Bluesky from time to time and there is (for me ofc) as much info/interesting stuff as I was getting from the other one before the acquisition.
it’s just people raging about trump and whatever brand they’re looking to try and cancel next.
it’s so far from the greatness of the original twitter. no tech community or content.
Although I still have a web/programming which I follow and have found some people interesting from Hackernews and others too in bluesky (emsh,simonw)
What is the HN consensus around lemmy? I really like lemmy and think that it might be better for tech stuff (almost similar to HN/reddit you can say and federated)
I used to follow lemmy c/technology but I do feel like HN is pretty unique in its own manner.
Regarding twitter alternative itself. Maybe mastodon too can be an alternative.
Another minor nitpick about bluesky is that its 200 characters limits actually really removes the tech community from too deep discussions imo. Although I guess twitter had that limit for long time too until it got removed but now I do see sometimes some tweets which are really long (sometimes even complete blog?)
It actually really (pissed?) me off so much that I ended up making a tampermonkey script which can actually write a long message automatically and split a message into 200 messages chunk and post them in a thread of sorts you can say although its very hacky and messy and it starts to glitch around 10 threads from what I remember.
I think there's a few tweetlonger-type services that people have tried to make, but with atproto they can at least embed that extra text into the post (100kb limit), so the site only needs to stick around to view it.
Woah I didn't know this was possible, definitely gonna try out leaflet.pub, I did know about it, let's see.
Thank you for telling me about this!
Edit: Now tried leaflet.pub, looks really cool. I am gonna use it from here on out because I used to use mataroa.blog or bearblog but they didn't have comments and comments were something I always wondered.
Its comments actually hook up to bluesky itself. this does feel interesting.
I might use it from here on out or maybe the fediverse alternative to something like this (i think its name was writty or something like this) but I am pretty sure that I am gonna nowadays write on leaflet too a litte haha!
Thanks for sharing me this. I knew about leaflet.pub but didn't know that it hooked with bsky so well. Is it a recent feature or is it the case where I maybe misremembering somethings?
I dunno. It probably depends on what you’re looking for.
It doesn't seem to be as bad anymore, a quick glance at the public feed suggests that the percentage of political posts has gone down, but considering how many times the word "toxic" appears in this linked blogpost, I'm guessing they're still banning anyone who expresses the "wrong" opinions, so the userbase is unlikely to grow much further in the future. It seems to have plateaued at around 1.2M daily likers.
Source for the stats: https://bsky.jazco.dev/stats
The US does not have a "far left" in any significant numbers, and never has. At least not in a self-aware sense.
Maybe you meant to say liberal, to which I'd agree.
That's not to say there isn't a "left" or "far left" on Bluesky, but there's no way it's a majority.
I agree echo chambers are a problem there, which is why I only posted there briefly before leaving. One feature that seemed to exacerbate the formation of echo chambers was users sharing and blindly trusting mass block lists to silence things they didn't want to hear (leftists and liberals alike).
Bluesky does, however. Clearly they've made that their target market, but that's also why growing beyond that base seems to be difficult for them,
Again, that's not to say lefists don't exist, but they are a tiny fraction, and always were a tiny fraction no matter what platform.
Don't rule out bots that exist in numbers to make the actual left appear like a deranged spectacle as a form of controlled opposition. Both parties of capital interests have a role in and benefit from these.
As for bots or “controlled opposition”: you don’t need conspiracy theories to explain why a heavily moderated platform with explicit cultural norms converges on a particular worldview. I’m disinclined to apply anything beyond Occam’s razor when accounting for “deranged spectacle” behavior; ordinary selection effects are sufficient.
This isn't in conflict with my original comment.
> As for bots or “controlled opposition”: you don’t need conspiracy theories to explain why a heavily moderated platform with explicit cultural norms converges on a particular worldview. I’m disinclined to apply anything beyond Occam’s razor when accounting for “deranged spectacle” behavior; ordinary selection effects are sufficient.
These aren't conspiracy theories, and they pre-date and extend beyond Bluesky. They are easily observable patterns in most modern news media and social media. For one, silos are much easier to advertise to. Follow the money, like everything else.
If anything, the mistake is treating the "reasonable", aspirational version as more real than the people who consistently show up, organize, and speak — and then assuming the most visible expressions must be "controlled opposition."
Once again, we seem to be in agreement on this.
> The underlying attitudes -- maximalism, moral absolutism, tolerance for disruption, and readiness to analogize opponents to historical evil -- are not artifacts of bots or manipulation. They’re characteristic features of a political subculture.
These things are not mutually exclusive. It's both, and people (and their bots) across the entire political spectrum are guilty of involvement.
There's really a problem that needs to be solved here. I really think anonymous or phony posting needs to stop. It's not helpful here. All it does is amplify false talking points with a "Fake it til you make it", "the loudest voice wins" methodology.
But unfortunately, engagement is financially incentivized now. So the big corps reap $$$$$ while the public burns itself down.
There is hope, but it requires enough people to care and act accordingly:
Can you explain what exactly you mean by "local nazis"? Are you getting ads for Nazi barber shops? Sieg Heil Heating & Cooling? Hitler Juice Bar and Bubble Tea?
If this was such a huge problem I'm sure we would have heard of it before.
Are they just disproportionately powerful then? Because the US does definitely have consistent far left trends and movements that overtake the mainstream. The OK hand gesture hysteria is maybe an evident example, but land acknowledgments? DEI statements? Fatphobia? Defund the police? All of these originate from far left positions.
You might be confused because several forces want you to be exactly that:
1) The right lumps/conflates everything from centrist liberal to far left as "the far radical left" with no in-between, which blurs many lines.
2) Center liberals who want a social media veneer they can feel good about will literally pose as leftists/marxists, but if you look at their other beliefs and behaviors (were they trying to sink Bernie, or not?) then it becomes immediately obvious they're ultimately loyal to the Dem party, and that means center liberals serving capital interests.
But I can't blame you or anyone else for falling for the above unless you've seen enough to know, like following both of Bernie's presidential runs and how he was systematically smeared by both liberals and their corporate media.
Identity politics / DEI / etc are a liberal obsession. Class politics is the focus of the actual far left.
...No? Bernie was super popular specifically with this audience. The more liberal people described themselves as, the more they supported Bernie: https://fivethirtyeight.com/features/what-defines-the-sander...
You can take the stance that nobody knows what any of these terms mean I guess, but then the picture gets kind of absurd, left-wing materialism loses all meaning, the church loses all relationship with the right, hell, from that standpoint Donald Trump campaigned as a leftist I guess? He did have a recurrent discourse around jobs and the working class.
It's not that they don't know. It's that they bend definitions to their advantage depending on what the context dictates. 538 is exactly the kind of outlet one would expect to do such a thing.
https://jacobin.com/2024/11/liberals-bernie-working-class-tr...
It seems the left-right spectrum serves better to confuse than to differentiate, and that the most productive discussions unfold when we talk issues instead.
I can moderate my own feed -- the majority of people don't need, want, or enjoy an overtly paternalistic hugbox, and especially if moderation tends to be unidirectionally skewed in one political direction. It's not surprising that growth is slow.
Lumping sophomoric image edits into that category is exactly the kind of moral and definitional inflation being actively used to manufacture pretext for suppressing speech under the guise of "moderation."
Harm has not been demonstrated. Annoyance or offense is not injury, and discomfort is not exploitation. Without evidence of systematic, material harm -- and without showing that enforcement would not introduce greater error and speech suppression -- the justification for moderation fails. Vague claims of illegality are irrelevant. A non-explicit image edit is not criminal in any US jurisdiction absent additional elements such as explicit sexual content, fraud, extortion, or targeted harassment; invoking “illegal in some states” -- without naming conduct and statute -- is just noise.
"Put them in a bikini" is closer to low-effort mockery than to any recognized category of sexual harm. The level of alarm being applied here is grossly disproportionate to the act itself and is merely being used as a pretext for broader intervention.
Ok, sure[1]. I don't know why I am, but I'm surprised people are going to bat for this. Where is your line, exactly? Is it legality? Is it further along?
[1] https://calmatters.org/economy/technology/2026/01/california...
>> The level of alarm being applied here is grossly disproportionate to the act itself and is merely being used as a pretext for broader intervention.
> I don't know why I am, but I'm surprised people are going to bat for this. Where is your line, exactly? Is it legality? Is it further along?
If something is illegal, that’s a clear boundary and the appropriate place for enforcement. If it’s legal, the burden is on anyone arguing for restriction to explain why speech controls are justified despite that -- what concrete harm exists, why existing law is insufficient, and why the remedy won’t create more error or suppression than it prevents.
Absent that showing, "this feels bad" or "this is alarming" isn’t a standard.
I'm not sure about that. I'd rather decide for myself what I want to read and what I do not. I'd love to not delegate this important decision to corporate overlords.
They also perform age checks because they are evil and complicit with obviously detrimental local laws, not because they want to protect the children.
If it makes you feel any better, it's more trivially bypassed than any website age-gate I've ever seen. (that's more complicated than a checkbox)
Of course they can comply and then in five years nothing more than cat pictures gets posted online.
(Not that this would be inherently bad)