Bluesky’s CEO meltdown: How leadership continues to fail its most marginalized users
tl;dr: Bluesky began with genuine promise: a new social media platform where users could feel safer, but inconsistent moderation quickly eroded trust. When systemically marginalized users pushed back, Bluesky’s leadership minimized concerns and began to target them for bans.
In October 2025, that tension culminated in a public meltdown by Bluesky’s CEO, who tried (and failed) to “post through it.” Today, Black and trans users continue to see their posts suppressed or removed, or their accounts banned for challenging Bluesky’s increasingly irresponsible leadership.
And, yes, I also explain the thing about the waffles.
For a shorter overview of the current controversy, cf. Anthony Ha, “Waffles eat Bluesky,” TechCrunch, 5 Oct 2025. It lacks context, but it’s a quicker read.
Background
I first pitched this story as a shorter, service-oriented article explaining the tensions behind Bluesky’s recent CEO meltdown. When it wasn’t picked up, I decided to write it here while the topic remains timely. Medium allows for a fuller exploration, but without an external editor, so please be kind with corrections; I’m happy to make them as needed.
Rather than just summarize the past few days, this piece traces the longer history of Bluesky’s failure to protect its most marginalized users, and situate the CEO’s recent meltdown in that context.
I have largely omitted names, because this piece isn’t about individual developers or users, it’s about a culture of systemic abuse and neglect. Responsibility for that culture ultimately rests with Bluesky’s CEO, Jay Graber.
Bluesky’s hopeful start
When Bluesky opened its beta in 2023, it was possible to be familiar with everyone on the app who posted semi-regularly. Bluesky genuinely seemed like a community of users interested in building a welcoming and safer online space.
I met many of my first mutuals through its infamous hellthread. Folks would wish each other good morning, share photos of themselves throughout the day, and sign off each night. The platform was so small that it was literally possible for a user to like every single post, thereby spreading small flashes of joy and the sense that, despite everything, we weren’t alone.
Yes, there was the usual infighting and occasional flameouts, but the introduction of the nuclear block often ended such conflicts quickly. I even wrote a widely-shared article extolling Bluesky’s culture of inclusivity and its growing list of features.
Bluesky’s development team felt responsive and engaged in the unique culture developing on Bluesky.
Bluesky cultivated systemically marginalized communities
During its closed beta, Bluesky was invite-only and those invites were coveted commodities. Early adopters were often people from systematically marginalized communities fleeing harassment and censorship on Twitter, Instagram, Facebook, and elsewhere. Influential users were sometimes given extra invites by Bluesky staff to help build a more diverse user base.
While the only user base Bluesky’s team would admit to cultivating was “writers with a Substack,” a newsletter platform notorious for hosting alt-right voices, hate speech, and libel, and whose co-founder would later defend monetizing neo-Nazi newsletters; in practice, Bluesky did in fact seek to expand its user base beyond the cis white tech bro early adopters.
One Black software engineer, for instance, received more than 500 invite codes and used them to bring hundreds of Black users onto the platform. When a user began harassing any trans accounts they could, Bluesky’s CEO took a stand against anti-trans harassment. And when Elon Musk erroneously declared the Latin prefix “cis” a slur on Twitter in June 2023, the Bluesky team gave everyone an additional invite code: a clear signal that if Twitter was no longer a safe space for Black people or trans people, Bluesky would be.
For a brief while, it worked, and Bluesky felt refreshingly different: more trans-friendly, more sex worker-friendly, and safer than most public social spaces online. This was a good thing. Everyone deserves spaces like that.
Many of Bluesky’s earliest tools (external apps, user lists, labellers, safety measures) were built by trans women and others trying to make the app a better place for everyone. (Some of these tools were later abandoned in the face of persistent transphobia on the app and Bluesky’s refusal to moderate it, or quietly removed by Bluesky’s opaque moderation team without notice or clear cause.)
One unaffiliated trans developer mapped Bluesky’s engagement networks, and it showed that the transgender shitposter cluster was among the platform’s largest and most active (though it seemed to capture only trans women with very few — if any — trans men or nonbinary people represented). Rolling Stone published an article celebrating Bluesky as a safer, more inclusive social media app.
During this period, Bluesky’s developers leaned into that image. One ordered cat ears and then skeeted a photo for Pride, shared a foot pic (a popular fetish catered to by Bluesky’s vibrant sex worker communities), and was friendly with “the dolls” to the point where it became increasingly uncomfortable.
Bluesky’s leadership deliberately courted systemically marginalized users and encouraged parasocial relationships with its lead staff. It seemed, for a moment, like this time things could be different.
But that sense of safety and mutual trust wouldn’t last long.
Bluesky’s anti-Black racism problem and the poster’s strike
Bluesky’s problem with anti-Black racism began early and remains ongoing.
The Black software engineer mentioned above who brought hundreds of Black users to the platform became a target of a racist harassment in May 2023. Despite the perpetrator’s clear history of racist abuse on the app, Bluesky’s moderation team failed to act. Rather than protecting the engineer or acknowledging the harm caused, Bluesky’s CEO dismissed her safety concerns, and responded by drafting a new policy to determine what qualified as a “death threat.” (That policy, notably, was later ignored when Bluesky banned users after an alt-right tabloid published a list of accounts that failed to express sufficient grief for the death of an alt-right podcaster who had himself wished death on many marginalized communities.)
It quickly became clear that users understood something Bluesky didn’t: Bluesky couldn’t code away social problems on the app. At some point, Bluesky’s leadership would have to do something about users who were openly racist, transphobic, misogynist, or otherwise abusive.
Nevertheless, the CEO’s solution was something she called “compostable moderation,” a skeet labelling system that users with the financial resources and technical know-how could develop and deploy on their own little corner of Bluesky. In practice, this was a way to outsource responsibility. Rather than Bluesky taking accountability for hate speech or death threats, users were expected to moderate themselves.
In July 2023, another controversy erupted. Users discovered that Bluesky’s username moderation filters failed to block racial slurs (most notably the n-word) and other harmful terms. When leadership, yet again, failed to take responsibility for this oversight or commit to a fix and instead went silent for 10 days. No more cute posts from developers, no more chummy back-and-forths with the CEO. In frustration, users turned to direct action.
The poster’s strike (or, more accurately, a boycott) with calls to action and a commitment to cease shitposting until Bluesky’s anti-Black racism was addressed. It’s not clear who initiated it, but it was widespread across the app. Users called for an apology and demanded that Bluesky hire a proper trust and safety team. More than two years later, the popular word cloud account Now Breezing continues to display a reminder to #ListenToBlackVoices.
Ironically, the utter silence from Bluesky’s team parallelled the poster’s boycott, and it seems to have been broken only by investor pressure. Some users received personalized apologies, but those appear not to have been sent under the directive of the company itself, but rather a newly hired employee.
While a general apology failed to emerge, the poster’s boycott did yield some results. It attracted media attention, and after more than a week of silence, Bluesky issued a response in which it promised to hire (or “expand”) a trust and safety team (though it was never clear whether one existed prior to the call for Bluesky to implement one).
Unfortunately, the person they hired to head that team in February 2024 proved to be a poor choice, as later events would show. Under his leadership, moderation remained inconsistent and opaque. Posts critical of Bluesky’s leadership were suppressed, and users were banned for speaking up.
Unlike other social media platforms, Bluesky still does not provide feedback when a report is filed, which is a small but telling omission that speaks to its broader culture and its unwillingness to take accountability for the decisions it makes.
Bluesky’s problem with persistent transphobia
Existing while trans on the Internet has never been easy. It’s difficult to find public spaces where you can simply be and post like anyone else without worrying that some bigot will try to dissect your identity or guess your chromosomes. For a while, despite some hiccups, Bluesky genuinely felt like one of those rare spaces.
That changed when the app moved from closed beta to full release. As waves of new users arrived from Twitter, they brought with them the platform’s notorious hostility with its culture of harassment, transphobia, and bad-faith engagement. An influx of new bots appeared with the sole purpose of sending death threats to trans people. I personally received more hate and death threats than I ever had before, for nothing more than including a trans flag in my profile, and that was on the lighter end of what others faced.
The full history of Bluesky’s problem moderating transphobia on its platform is longer than I can cover here, but a key episode helps explain the context of the CEO’s recent meltdown. Late last year, a transphobic opinion writer joined Bluesky and was swiftly banned. With his history of writing anti-trans propaganda that has been used in US court systems to roll back trans rights, sharing information from doxxing sites, and picking personal fights with any trans person he comes across, it makes sense that he would be banned. Yet, after publicly complaining about it elsewhere and a chat with Bluesky’s head of trust and safety, he was reinstated. Since returning, he’s done little but gloat, spread transphobic misinformation, and continues to harass trans users, including evading blocks to dox Bluesky users on his podcast. He quickly became Bluesky’s most-blocked user.
In response, the community did what it’s learned to do: organize. Users reported his posts en masse, petitioned the head of trust and safety, and eventually appealed directly to the CEO who had once been seen as an ally to trans users. The controversy received a ton of media attention. None of it worked.
A petition with more than 28,000 signatures — including Lizzo’s! — calling for the writer’s removal was sent to Bluesky’s leadership. It received no reply. He remains active on the platform, continuing to post misinformation that fuels hostility toward trans people and contributes to the broader rollback of trans rights in the US.
The leadership’s ongoing silence led one trans woman to create a bot account called “What’s our head of T&S doing?” that automatically logged every post liked by Bluesky’s head of trust and safety. When the bot revealed that he had liked explicit teen-themed bot porn content using his professional account, the bot was swiftly deleted and the trans woman who ran it was suspended without warning or recourse. Her account hadn’t violated Bluesky’s terms of service; it had simply embarrassed someone in power.
In June 2025, the new US Vice President joined Bluesky, explicitly for the purpose of trolling trans people about the US Supreme Court’s decision to block healthcare for trans youth. His account quickly became the most blocked on the platform, surpassing even that of the transphobic opinion writer, who preened about it.
Bluesky’s selective moderation
This article is already long, but it’s worth noting that Bluesky’s moderation failures extend beyond racism and transphobia. The platform has repeatedly banned Palestinian accounts seeking aid during an ongoing genocide. These actions led to both an open letter and a petition condemning Bluesky’s behaviour. Leadership dismissed both, despite years of concern from users and mounting negative press.
Bluesky’s automated image-censorship system also shows a clear bias. It over-polices bodies the algorithm reads as female, often flagging or removing innocuous images, while allowing similar or far more revealing content from accounts it reads as male. Sex workers have been especially targeted by this uneven enforcement.
In short, Bluesky’s moderation is not neutral; it is selective. The platform is quick to silence, censor, or ban users it finds inconvenient or embarrassing, yet consistently unwilling to act against hate speech, misinformation, or other antisocial behaviour.
Bluesky’s CEO meltdown
This brings us to Wednesday, 1 October 2025. Bluesky’s CEO quote-skeeted a post from an account often featured in the company’s promotional materials. The skeet referenced a popular meme from Twitter about off-topic replies. A trans user replied, raising a legitimate concern: that the transphobic opinion writer still posed an active threat to user safety on the platform.
The CEO’s response was an all-caps “WAFFLES,” a dismissive and misguided attempt at humour that effectively declared the topic off-limits. But user safety is never off-topic, especially when leadership has repeatedly failed to protect its most vulnerable users. The post was immediately ratioed, as trans users and allies voiced their frustration at yet another flippant, tone-deaf response.
The next day, the CEO doubled down, posting a photo of waffles, openly trolling trans users who had criticized her.
After a period of quiet on the app, the transphobic opinion writer returned and followed her lead, skeeting his own smug photo of waffles in apparent solidarity with the CEO who so steadfastly defends his presence on the platform.
The message is clear: Bluesky’s leadership isn’t just tolerating transphobia: it is indulging it.
On day three, Friday, 3 October 2025, she escalated again, mocking a user by suggesting another poster’s strike, similar to the one that had successfully drawn attention to Bluesky’s anti-Black racism problem. The CEO implied that such action was pointless. But it wasn’t pointless: the first strike had forced investor intervention and prompted Bluesky’s leadership to issue a public statement.
Curiously, the first two posts that sparked the backlash have since been suppressed from the CEO’s profile, though they remain visible through search. This quiet erasure fits a familiar pattern of attempting to move on without accountability or engagement. And, as before, users who challenge leadership’s behaviour have faced retaliatory moderation and suppression.
Bluesky’s CEO skeeted, “Harassing the mods into banning someone has never worked. And harassing people in general has never changed their mind.” But is it harassment to ask a social media platform’s leadership team to ensure user safety, especially when they consistently refuse to do so?
As one user pointed out, customers threatening to cancel their service often do get results. (Indeed, we saw this recently with Disney and Jimmy Kimmel.)
The CEO’s response was again flippant, “Are you paying us? Where?”
I replied that Bluesky’s users are its product, because “Without us, Bluesky is just code on a server.” The post quickly gained hundreds of likes and reskeets before being quietly suppressed, along with several other measured responses.
My reply no longer appears in the thread, though it’s still visible via direct link or reskeet. It’s hard not to see this as an effort to deflect attention from the obvious truth: the value of any social platform lies in its users and Bluesky’s leadership resents that. Perhaps the CEO was embarrassed by the ratio, especially after screenshots began circulating on Reddit and across Bluesky itself.
One Bluesky developer then screenshot my post, stripped it of context, and dunked on it without alt text, making it less accessible to disabled users like me. Interestingly, a significant portion of the replies agreed with me. Users understand that if we’re not paying, we are what’s valuable about the social media platform: our engagement, our content, and our attention are what Bluesky uses to raise capital and prove its worth as a product.
As one user observed, if the CEO insists Bluesky’s users aren’t customers, and a developer claims we aren’t the product. What, exactly, are we?
Most of the replies still visible under the CEO’s original post are overwhelmingly hostile and belligerent. It’s difficult not to wonder if this selective visibility is meant to manufacture sympathy for a “poor beleaguered CEO” rather than acknowledge that, yes, on a social media app, users are what make it social.
Many have urged her to hire a PR team. Others joked it’s more entertaining to watch her continue to shit the bed.
Still, Bluesky’s community is responding in the way it always has: organizing, documenting, and holding power to account. Users have circulated a thread listing current investors for anyone who wishes to contact them directly. If that thread disappears, here are a few of the investors named so far:
- DarkMode Ventures
- Skyseed
- Alumni Ventures
- Greylock VC, “Empathy is the foundation of our work.”
- Blockchain Capital
Reasoned appeals have rarely worked with Bluesky’s leadership. Direct action has.
Update: With apparently no sense of irony, the app’s official account announced today that the platform had hit 2 billion skeets. Who wrote those skeets? And what shareholder value do they represent, if not the very people Bluesky’s leadership insist are neither customers nor the product?
Bluesky suppressing skeets critical of its CEO and banning established community builders
At some point, Bluesky’s CEO must have taken someone’s advice to stop trying to post through the backlash and shift tactics: from baiting trans users directly to quietly shadowbanning and permabanning those who raised concerns.
My reply that ratioed the CEO was the first I noticed to be suppressed, but it wasn’t the last. Others quickly followed. I could only track those from people I follow or know of, but a clear and troubling pattern quickly emerged.
On Sunday, 5 October 2025, a popular trans comic artist had one of her skeets critical of Bluesky’s CEO removed. Code analysis later confirmed that it was indeed taken down, though it has since been restored.
That same day, Bluesky permanently banned a Black space lawyer, social justice advocate, and long-time community builder without warning or cause. The likely reason? His criticism of the CEO’s recent decision to follow a well-known transphobe flagged by the Follow Tracker labeller. A justification was manufactured after the fact based on a two-week-old skeet that did not violate Bluesky’s terms of service.
Users are understandably alarmed. This fits into a broader and persistent pattern: retaliatory moderation, lack of transparency, and the continuation of anti-Black and anti-trans bias at the highest levels of Bluesky’s leadership.
The result is predictable: a deepening loss of trust, and still, no accountability.
Update: It has since emerged that the community builder’s account has not only been permabanned from Bluesky, but also blocked from WhiteWind, Skylight, Bluescreen, Flashes, and “basically all ATproto apps/projects.” This effectively negates the very premise of the AT Protocol’s federated model.
Second update: A Jewish science journalist was briefly suspended after publicly criticizing Bluesky’s CEO. A weeks-old skeet condemning Christian Zionism was presented as justfication, but this is clearly manufactured pretext.
Third update: The Follow Tracker labeller — the tool that showed who was following nine prominent racists, transphobes, and other figures popular with the US alt-right — is shutting down. Its operator insists this isn’t because of Bluesky’s CEO (and thus, presumably, not because the labeller identified her as following two well known anti-trans activists), but rather “because some of you can’t be trusted to use this information reasonably and responsibly,” which is a curious justification coming from an account literally named @for.shame.wtf.
At this point, all pretense is gone. This is outright retaliation by Bluesky’s leadership against any who dare criticize it.
Conclusion
I’m fully aware that I’m not important. I’m just a random trans person from Tkaronto, a historian trained in social and cultural history, and that perspective has shaped how I approached this piece.
I’m not owed anything, and I’m no one to Jay Graber or Bluesky. But that’s precisely the point: I’m like every other faceless user Bluesky’s leadership prefers to abstract and ignore.
In the wake of the current US president’s second inauguration and his administration’s openly transphobic, anti-DEIA policies, Bluesky initially stood apart. It was one of the few major platforms that didn’t rush to rewrite its terms of service to allow slurs against queer and trans people, unlike Twitter, Google, or Meta’s suite of apps (Facebook, Instagram, WhatsApp, Threads).
It is notable that the first two social media instances built on the AT Protocol outside of Bluesky are Blacksky and Northsky Social. Blacksky, launched earlier this year along with a separate instance for non-Black users, was created in response to persistent anti-Black racism on Bluesky. Northsky, a transgender-focused instance, is similarly being developed as a space for users who no longer feel safe on Bluesky due to ongoing harassment, transphobia, and selective moderation. Both platforms reflect a broader trend: when leadership fails to protect its marginalized communities, users take matters into their own hands, creating spaces that prioritize safety, accountability, and inclusivity — values that Bluesky’s leadership once claimed, but repeatedly neglectes.
Bluesky’s early promise has eroded. The platform’s inconsistent and retaliatory moderation practices have drifted steadily rightward, protecting prominent racists and transphobes while suppressing those who speak out against them. Even mild criticism of the company’s failures to safeguard marginalized users is now met with shadowbans, takedowns, or silence.
That’s not leadership. That’s an abdication of responsibility — and we deserve better.
Further reading
- Aouir, Reem. “Bluesky urged to review ‘punitive’ policies towards Palestine content.” Middle East Eye, 26 May 2025.
- Bhuiyan, Johana. “Social media accounts of Palestinians desperate for funds are being flagged as spam.” The Guardian, 8 August 2025.
- Chen, Caiwei. “Bluesky users are cheering their exodus from X — the switch is leaving Palestinian posters behind.” Daily Dot, 19 November 2024.
- Collier, Kevin. “Bluesky’s growing pains.” NBC News, 22 December 2024.
- ElizaBeth. “BlueSky Staff Refuse to Ban Jesse Singal.” Medium, 4 October 2025.
- GLAAD Accountability Project. “Jesse Signal.” GLAAD, 21 April 2023.
- Graber, Jay. “Composable Moderation.” Bluesky Blog, 13 April 2023.
- Ha, Anthony. “Waffles eat Bluesky,” TechCrunch, 5 October 2025.
- Hendrix, Justin. “Substack Cofounder Defends Commercial Relationships with Nazis.” Tech Policy, 21 December 2023.
- Jeong, Sarah. “Bluesky showed everyone’s ass.” The Verge, 2 May 2023.
- Keenan, Sara. “Is Twitter Competitor Bluesky Failing To Protect Its Black Users?” People of Color in Tech, 12 June 2023.
- Kirkland, Colin. “Bluesky’s Safe-Space Reputation Threatened By One User’s Views On Trans Issues.” MediaDailyNews, 13 December 2024.
- Klee, Miles. “‘It’s a Huge Relief’: Trans ‘Shitposters’ on Bluesky Feel Safer Away From Twitter.” Rolling Stone, 3 May 2023.
- Klee, Miles. “J.D. Vance Joins Bluesky, Immediately Becomes Most Blocked Account on App.” 19 June 2025.
- Malcolm, Jeremy. “Bluesky melts down over Jesse Singal.” 16 December 2024.
- Mara-McKay, Nico. “How to Bluesky.” Medium, 16 August 2023.
- Messman-Rucker, Ariel. “Controversy brews on Bluesky after users complain about anti-trans podcaster.” Advocate, 13 December 2024.
- Newitz, Annalee. “Bluesky is just another Twitter clone and that isn’t a good thing.” NewScientist, 7 June 2023.
- Owen, Greg. “JD Vance joined Bluesky & posted an anti-trans message. Users immediately paid him back.” LGBTQ Nation, 20 June 2025.
- Perez, Sarah. “Bluesky invites become a hot commodity as demand for the Twitter alternative outstrips access.” TechCrunch, 2 May 2023.
- Perez, Sarah. “Bluesky at a crossroads as users petition to ban Jesse Singal over anti-trans views, harassment.” TechCrunch, 13 December 2024.
- Plummer, Kate. “JD Vance Becomes Most Blocked Person on BlueSky Days After Joining App.” Newsweek, 20 June 2025.
- Robison, Kylie. “Inside Twitter rival Bluesky’s first major crisis, as investors pressured CEO Jay Graber to speak out about racist incident.” Fortune, 31 July 2023.
- Serano, Julia. “My Jesse Singal story.” Whipping Girl, 5 December 2017.
- Sung, Morgan. “For Bluesky to thrive, it needs sex workers and Black Twitter.” TechCrunch, 2 May 2023.
- Sung, Morgan. “Bluesky’s growing pains strain its relationship with Black users.” TechCrunch, 8 June 2023.
- Sung, Morga. “Bluesky sends some users personalized apologies after racism controversy.” TechCrunch, 27 July 2025.
- Tolentino, Daysia and JJ McCorvey. “Black Tech Twitter, trans users and more marginalized groups flock to Bluesky.” NBC News, 4 May 2023.
