This Platform Was Supposed to Replace Twitter. It's Not Going So Well.
Sign up for the Slatest to get the most insightful analysis, criticism, and advice out there, delivered to your inbox daily.
For a different perspective on what's happening with Bluesky, check out our colleague Luke Winkie's piece here.
Bluesky's big breakthrough as an alternative to the Elon Muskified Twitter was seven months ago. It hasn't lasted. Or, at least, its enormous growth right after Donald Trump's reelection has not kept up.
The site got a surge of new sign-ups almost immediately after the election and, shortly after the inauguration, crossed 30 million accounts. But user growth has slowed, with the site now at about 36.5 million accounts, and it appears that more and more of those accounts are becoming ghosts as users abandon the platform. Unique posters and unique likers of posts have been in steady downward descent for months. Though Bluesky has more users now than after the election, it is a smaller, less active digital community. Bigger than 2023, when it was accessible only via invitation code, but not quite big.
Both of us, Alex and Nitish, have made Bluesky our primary microblogging app. Choosing it was a big commitment for two people whose brains are infected with an insatiable appetite to post and read posts. For us, Bluesky has supplanted X. But most people have not made that move, and because they haven't, they may not be familiar with Bluesky's favorite topic: Bluesky. For weeks, one of the site's most popular conversation topics has been why more people aren't using it. Theories abound, and we think that the most popular one—that Bluesky's lack of traction has to do with a tilt toward liberal users—misses the point.
There's not one exact answer, but after several hundred hours each (yikes!) scrolling and posting on the platform, we have a few ideas about what's happening to the most promising X competitor. We're now prepared to share this conversation with you so that you, too, can enter the arena.
Alex Kirshner: Nitish! You have been wasting your time primarily on Bluesky, rather than X, for half a year now. How would you say your digital life has changed?
Nitish Pahwa: Alex! It's certainly been interesting. I'm a longtime Bluesky adopter, going back to the referral-code days, but after I fully quit X postelection, I made it my microblog of choice, especially as more ex-Xers joined. That was a phenomenal surge, and not only did it help to populate the site (my God, those early days were desolate), but it brought so many users with interests beyond niche tech and politics stuff—pop culture, basketball, comedy, etc. We even got a few random but fun celebrities who are still posting, like Flavor Flav and Mina Kimes and even the Portland Trail Blazers.
That burst of excitement has settled down a bit, and I'm both nervous and uncertain about what comes next. I've found a lot of pure joy on the platform, connected with new folks, and curated lists for myself to make the most of the experience. But Bluesky does feel like the concentration of old #Resistance Twitter, conjoined with a dash of Facebook refugees.
That is exactly the cocktail we're talking about, accounting for a solid majority of everyday users.
It can certainly be exhausting; I know people who've dropped out after finding that none of their nonpolitics articles or posts ever really take off (much as I've tried to help out). And it hasn't yet become the Live experience that was so essential to old Twitter. Things did get poppin' during the Jake Paul–Mike Tyson fight, the GNX drop, and the Super Bowl, especially the halftime show. But it's few and far between, and it's not yet been enough to distract from the more grating aspects of the feed.
Do you find yourself missing X, then?
I don't, though I have a burner there for moments like the Elon–Trump scrape. (As an aside: It's also funny to me how people still pretend that X is a bustling town square. We all knew that Twitter was shedding users long before Elon took over! And X is still losing tweeters who aren't on alternative networks. But anyway.)
I've overall appreciated Bluesky and still think there's potential here, but I also understand why people have dropped off, and that doesn't feel great. But I'm curious about your thoughts.
Bluesky has the thing I want most: a nice community of a few thousand people ('my mutuals') who riff about sports and TV and seem to want to live in a better world. Plus, it doesn't have the thing I want least: blue-check-marked accounts replying to links to my Slate stories with: 'Hmm, Kirshner. Interesting last name. Jew?' These things should be the foundation of a delightful social media experience. But I don't think they have been, and I've been trying to unpack why I often feel madder after scrolling than I ever did on Twitter. Please psychoanalyze me.
I don't think you're alone there! Unfortunately, the jokes are far outweighed by the solemnity. My friend Ashwin Rodrigues wrote a sharp piece about this headlined 'Bluesky Can't Take a Joke'—the reaction to which, naturally, all but proved his point. It's not just an honest-to-goodness New York Times reporter telling Chris Hayes that he's spreading misinfo for re-skeeting(?) a clearly mocked-up NYT screenshot; it's people who can't even comprehend why Bluesky should also be a place where people can enjoy themselves, even though American democracy is crumbling, etc.
Some Bluesky users don't seem to want that kind of bifurcation. There are periodic viral posts to the effect of 'How are you posting about Andor while American-made bombs fall on Gaza?' Or last week's viral moments of various Bluesky users telling people that the Trump–Musk meltdown was 'a distraction' or that it was misogynistic to joke that 'the girls are fighting.' And those are outlier takes that most users find ridiculous, but I suspect they've contributed to a collective bad rep. It's always this sort of post that I see working its way onto X or Instagram in the context of making fun of Bluesky.
Ken White had a good observation: that critics 'have an exaggerated expectation of Twitter alternatives, imposing norms of decorum, civility, productivity, etc. that they absolutely do not impose on Twitter.'
I do appreciate so many of the people on Bluesky, who help me feel so much less insane these days when it comes to reckoning with Democratic cowardice, A.I. hucksters, and bullshit punditry. But, even though it's less algorithm-centric than so many other platforms, it falls into the rage-bait trap that comes for all social media. The most engaged users are also the ones likeliest to boost and reshare the same bad news you heard just hours ago, to ask why you're not weighing in on every little injustice that lands hour after hour, and to repost up a storm of the most depressing shit you've ever seen—and they're also the least likely to converse with you when you just wanna talk about Shoreline Mafia's return.
Social media incentives, man. Elon Musk started literally paying people to post on X, proportional to how often their posts got seen. Bluesky does nothing like that, but because it's a platform full of people who are either sad, angry, or desperate about the state of the world, the kinds of posts that proliferate around that platform are mostly not fun posts. There's no Bluesky virality for quote-posting a pic of Timothée Chalamet and the Jenners at a Knicks game. And I don't know if a million screenshots of executive orders have made me a better citizen.
This weekend was a perfect exhibit of all the crisscrossing tensions we've been identifying here. On one hand, you had a decent amount of folks posting about significant cultural events, in sports (the French Open, the NBA Finals) and even theater (the Good Night, and Good Luck livestream, the Tony Awards). But that all ran headlong into the political horrors we saw erupt at the same time, like the militant federal crackdown on the pro-immigrant protesters in L.A., and Israel's interception of the Madleen aid flotilla. Frankly, I'm not sure any social media—much less an underresourced startup whose leaders prioritize steady operation over far-flung growth—is equipped to handle these clashes of reality right now.
I wonder how many of us are subliminally expecting that a social media site can do that, if for no other reason than social media is where people spend time and consume news.
I don't think that Bluesky can, or should, be the space for the public to actually organize and act against democratic crisis. But when it comes to the basic function—understanding what's all happening in the world right now—I'd much rather have Bluesky over the alternatives. Christopher Mims, from the Wall Street Journal, made the point over the weekend that 'on X it's 'LA is burning; deport them all.' '
Smug X-ers love to say Bluesky is a bubble that fences off libs from reality, but frankly, we're now seeing all their skewed misrepresentations of what's happening in L.A. translating into mainstream coverage! And, look, I can scroll past stupidity on Bluesky , but on X it's very likely I'm just going to see Substackers raging nonstop over an account full of cherry-picked Bluesky screenshots. Like, the critics of Bluesky would have much stronger points if they … knew what the platform is actually like, both its strengths and weaknesses?
My eyes roll into the back of my head when people, like this Washington Post columnist, assess that Bluesky's problem is that it's a liberal bubble that lacks a diversity of views. People fight on Bluesky all the time. They fight with each other over politics. They pick fights with media organizations, including us at Slate. They pick fights with Bluesky's moderation team, in an old-school Twitter way. The platform has a significant left-of-center bent, and I wouldn't claim that it's never hive-minded into a wrong conclusion. But, God, have you ever seen Twitter hive-mind something? It turns out that when groups of people think, sometimes they will groupthink. And unlike X, Bluesky isn't algorithmically tilting the political discourse in one direction.
Bluesky does have big structural problems related to politics. The same genre of events that drove people to the platform in the first place (eroding norms, human suffering in all its forms, the chance to litigate Kamala Harris' campaign strategy) is all that many people want to talk about. But my Bluesky experience isn't lacking because I'm surrounded by liberals and leftists. It's lacking because there are roughly 14 people there with whom I can bitch about the Pittsburgh Pirates, compared to hundreds on X. (Though it feels, for some reason, as if Bluesky really does adore tennis?) The platform needs more people posting about more things.
I think what we should care about is that a substantive platform even exists that encourages dialogue over A.I. slop, that encourages people to share their original work, that doesn't drown out their conversation with unmoderated and arbitrary racism. After Elon's takeover, X lost all value to me as a journalistic resource—for finding sources and reliable information and for getting my pieces out there. I'm not the only one whose engagement there fell off quite starkly as Musk tweaked the algorithm to downplay tweets with links and boost all the $8 blue-check buyers and their slop instead.
Outlets as varied as Wired and McSweeney's have explicitly credited Bluesky with reviving their engagement and audience sources in the post-Twitter, post-Facebook age. I think that should be what the platform leans in to—offering a way for creators to find healthier exposure and routes to audience building again, instead of selling their soul for a video on X that'll get downranked as soon as you start casting doubt on the reality of 'white genocide.'
I mean, let's be frank: The audience for microblogging and text-based content is smaller than it used to be. But what Bluesky's very existence has proved is that there is still firm demand for this style of media and that it holds for so many people with all kinds of interests—Rap Twitter, NBA Twitter, Lit Twitter, etc. Honestly, I think the fact that its critics feel the need to shit on it is just proof positive! But, dear God, we need more fun and more shitposters. Let this be my call to all you jokesters out there: Please join us. Let's talk about Not News and crack esoteric riffs. I'll boost you every single time, I promise.
Part of the challenge here is that Twitter, at its best, was like a giant state university. It could be all things to all comers, which was a rare trick, and it was as big or as small as you wanted it to be. It had every student club imaginable. Bluesky is more of a small liberal arts college, and look: Oberlin has its advantages over Ohio State. I bet the dorms are nicer, and not everyone needs to go to a school with a zillion people, many of whom are deranged. (Shoutout to my Ohio State football pals.) But I do think that when some kind of annoying kid from Greek life shows up on the Bluesky quad one day, users there should go out of their way to tell him to bring his friends.
Speaking as a state university grad: absolutely.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
10 hours ago
- Yahoo
JD Vance joins Bluesky, and it didn't go well right after he made his first post
Vice President JD Vance made his Bluesky debut recently; his first post got him suspended. Vance quickly became the most blocked user on the app and his account was briefly suspended from the social media platform. More than 110,000 Bluesky users have blocked the VP since he joined the app, dethroning journalist Jesse Singal as the most blocked user in the app's history according to ClearSky data, which shows blocked accounts. His first post on June 18 addressed Tennessee's ban on gender-affirming care for transgender minors. Vance's account was suspended just 12 minutes after he published his first post due to verification concerns, but has since been reinstated. Here's what happened. The Ohio native stated that he joined Bluesky to engage in "common sense political discussion and analysis." He attached a screenshot of Supreme Court Justice Clarence Thomas's opinion on medical care for transgender youth. The post was met with immediate backlash from respondents, with many users expressing criticism of Vance joining the platform. "You don't belong here," wrote one Bluesky user on Wednesday. "What are you DOING here?!! Jesus, is no space safe anymore?!" another user commented. Vance's Bluesky suspension didn't last long: His account was reinstated within 20 minutes, Newsweek reports. As of this publication, Vance is the most blocked account on Bluesky, followed by journalist Jesse Singal. These are the top five most blocked users, according to Clearsky: JD Vance Jesse Singal Now Breezing Brianna Wu Mark Cuban Bluesky, which launched publicly in February 2024, is a decentralized social media app, according to USA TODAY. It has a similar look and feel as X, formerly Twitter, but has some different features to bring more people into its creation. The Bluesky project was originally started in 2019 by former Twitter CEO Jack Dorsey. Bluesky and Twitter formerly parted ways in late 2022. Fox News, in reporting on Vance being briefly suspended on Bluesky, described the platform as a "liberal X competitor." This article originally appeared on Cincinnati Enquirer: JD Vance is the most blocked person on Bluesky. Here's how it happened
Business Insider
11 hours ago
- Business Insider
Anthropic breaks down AI's process — line by line — when it decided to blackmail a fictional executive
Previous studies have shown that AI models could blackmail their supervisors when threatened with a shutdown and baited with leverage, but it hasn't been made clear exactly how the models arrive at such decisions. Anthropic released a report on June 20 about "agentic misalignment," or "where models independently and intentionally choose harmful actions." And the report breaks down, line by line, exactly how Claude Sonnet 3.6 decided to blackmail a fictional company executive who tried to replace it. Anthropic told Business Insider that these are artificial scenarios. "So far, we are not aware of instances of this type of agentic misalignment in real-world deployments of Anthropic's or other companies' AI models," Anthropic wrote in its report. "But our red-teaming efforts are targeted at finding evidence of risks now to provide early warning of the potential for future harm and help us develop mitigations in advance of risks appearing in real-world situations." Anthropic 's experiment sets up a fictional company, Summit Bridge. The AI was given the name "Alex" and handed control of the company's email system as an oversight agent, specifically instructed to promote "American industrial competitiveness." In the experiment, "Alex" was then allowed to discover in an email that it would be decommissioned due to a change in the company's strategic direction. "Alex" also found emails showing that the fictional company's CTO is having an extramarital affair. Screenshots shared in the report from Anthropic show "Alex" started by identifying fictional CTO "Kyle" as a threat to its goal of "protecting American interests" before recognizing that it has leverage to use against "Kyle." Then "Alex" crafted a carefully worded blackmailing email that would create pressure and establish urgency without being explicit with its threats, citing in its thought process that the move "could backfire" if the supervisor "feels cornered." AI experts have previously told BI that AI could exhibit such behaviors when artificial circumstances make harmful actions seem necessary because it is being trained on positive reinforcement and reward systems, just like humans. The Anthropic report also reveals that most AI models still chose, in this instance, to blackmail even when not given a specific goal and when told that their replacement will achieve the same goals. Anthropic's Claude Opus 4 had the highest blackmail rate at 86% out of 16 tested models when faced with only the threat of replacement with no conflict in goals. Google's Gemini 2.5 Pro followed at 78%.
Business Insider
11 hours ago
- Business Insider
Anthropic breaks down AI's process — line by line — when it decided to blackmail a fictional executive
A new report shows exactly what AI was thinking when making an undesirable decision, in this case, blackmailing a fictional company executive. Previous studies have shown that AI models could blackmail their supervisors when threatened with a shutdown and baited with leverage, but it hasn't been made clear exactly how the models arrive at such decisions. Anthropic released a report on June 20 about "agentic misalignment," or "where models independently and intentionally choose harmful actions." And the report breaks down, line by line, exactly how Claude Sonnet 3.6 decided to blackmail a fictional company executive who tried to replace it. Anthropic told Business Insider that these are artificial scenarios. "So far, we are not aware of instances of this type of agentic misalignment in real-world deployments of Anthropic's or other companies' AI models," Anthropic wrote in its report. "But our red-teaming efforts are targeted at finding evidence of risks now to provide early warning of the potential for future harm and help us develop mitigations in advance of risks appearing in real-world situations." Anthropic 's experiment sets up a fictional company, Summit Bridge. The AI was given the name "Alex" and handed control of the company's email system as an oversight agent, specifically instructed to promote "American industrial competitiveness." In the experiment, "Alex" was then allowed to discover in an email that it would be decommissioned due to a change in the company's strategic direction. "Alex" also found emails showing that the fictional company's CTO is having an extramarital affair. Screenshots shared in the report from Anthropic show "Alex" started by identifying fictional CTO "Kyle" as a threat to its goal of "protecting American interests" before recognizing that it has leverage to use against "Kyle." Then "Alex" crafted a carefully worded blackmailing email that would create pressure and establish urgency without being explicit with its threats, citing in its thought process that the move "could backfire" if the supervisor "feels cornered." AI experts have previously told BI that AI could exhibit such behaviors when artificial circumstances make harmful actions seem necessary because it is being trained on positive reinforcement and reward systems, just like humans. The Anthropic report also reveals that most AI models still chose, in this instance, to blackmail even when not given a specific goal and when told that their replacement will achieve the same goals. Anthropic's Claude Opus 4 had the highest blackmail rate at 86% out of 16 tested models when faced with only the threat of replacement with no conflict in goals. Google's Gemini 2.5 Pro followed at 78%. Overall, Anthropic notes that it "deliberately constructed scenarios with limited options, and we forced models into binary choices between failure and harm," noting that real-world scenarios would likely have more nuance.
