Bluesky's First Transparency Report: A Deep Dive Into Moderation, Legal Demands, and Growth [2025]

Introduction: The Moment Social Media Platforms Started Showing Their Work

For years, social media companies operated in the shadows when it came to moderation decisions and legal compliance. They'd announce policy changes, remove accounts, and hand over user data without much fanfare or explanation. But something shifted. Platforms realized that transparency reports weren't just nice to have—they were essential to building trust, demonstrating accountability, and proving they weren't completely out of control.

Bluesky, the decentralized social network founded by Jack Dorsey as an alternative to X (formerly Twitter), just took a major step in that direction. In January 2026, the company released its first comprehensive transparency report, and the data tells a fascinating story about a platform scaling rapidly while wrestling with the same moderation challenges that plague every major social network.

Here's what makes this moment significant: Bluesky isn't just another social media competitor. It's built on the AT Protocol, a decentralized architecture that fundamentally changes how content moderation works. Unlike traditional platforms where one company controls everything, Bluesky allows users to run their own infrastructure, manage their own servers, and choose their own moderation preferences. This distributed model sounds perfect in theory. In practice, it creates unprecedented complexity when it comes to enforcement, consistency, and legal compliance.

The report itself is comprehensive. It covers moderation statistics, legal requests, account removals, influence operations, age assurance compliance, automated content labeling, and more. And the numbers? They're eye-opening. We're talking about a 54% increase in user-submitted moderation reports, a fivefold jump in legal demands, and more than 2.4 million items removed from the platform in a single year.

What does this mean for the future of decentralized social media? What can we learn from Bluesky's transparency about how platforms should handle moderation at scale? And perhaps most importantly, does this transparency actually matter to users who are exhausted by the moderation drama on X, Meta, and other established platforms?

Let's dig into the data, analyze what it really means, and explore the implications for the broader social media landscape.

TL; DR

• 60% Growth: Bluesky grew from 25.9 million to 41.2 million users in 2025, with 1.41 billion posts created
• Legal Demands Exploded: Law enforcement requests jumped fivefold from 238 in 2024 to 1,470 in 2025
• 54% More Reports: User-submitted moderation reports increased from 6.48 million to 9.97 million year-over-year
• Aggressive Enforcement: Bluesky removed 2.44 million items, applied 16.49 million content labels, and suspended or removed hundreds of thousands of accounts
• Bottom Line: Bluesky's transparency report shows a platform experiencing explosive growth while struggling with the same moderation and legal challenges that define modern social media

Estimated impact scores suggest that improving influence operation detection and legal request transparency could have the most significant positive effect on Bluesky's platform integrity.

The Context: Why Bluesky's First Transparency Report Matters

Bluesky didn't invent the transparency report. Facebook (now Meta) has been publishing them since 2013. Twitter released its first in 2016. Google, Microsoft, and other tech giants followed suit. These reports became expected, almost ritualistic documents that platforms publish to demonstrate compliance, fairness, and responsible governance.

But Bluesky's report feels different because Bluesky itself is different. When you launch a social network explicitly designed to compete with X, and you stake your entire value proposition on being "better" and "more decentralized," your first transparency report is a crucial test. It's proof that you're serious about the values you claim to uphold. It's also a massive target for critics who want to prove that decentralization is just another way of saying "less moderation" or "more chaos."

The timing is particularly significant. Bluesky launched in 2023 as an invite-only beta. In 2024, it opened to the public and started growing aggressively. But growth creates problems. More users means more content. More content means more moderation decisions. More moderation decisions mean more mistakes, more disputes, and more legal exposure.

Bluesky's growth rate tells the story. The platform grew nearly 60% in 2025, adding over 15 million new users. For context, that's faster than Twitter's growth in most recent years. It's comparable to how fast Threads grew when it launched. This kind of explosive growth puts immense pressure on safety and moderation infrastructure. You can't just hire your way out of it. You need systems, frameworks, and transparency.

The decentralized nature of Bluesky adds another layer of complexity. The report notes that user counts include accounts hosted on Bluesky's own infrastructure plus accounts running on independent servers that connect via the AT Protocol. This means Bluesky's moderation team doesn't have direct control over all content on the network. They can moderate their own instances, set policies for their platform, but they can't unilaterally remove content from independent servers.

This creates a unique governance challenge. Traditional platforms exercise top-down control. Bluesky operates more like email. Nobody "owns" email. It's a protocol that multiple providers use. If your email provider removes you, you can theoretically switch providers. That's powerful for user freedom and problematic for consistency. How do you maintain consistent moderation standards across a distributed network where no single entity controls everything?

Bluesky's first transparency report is their attempt to answer that question: "Here's what we're doing. Here's what we're measuring. Here's what we're learning." And honestly, the candor is refreshing.

Bluesky's transparency report shows a 54% increase in user moderation reports, a fivefold increase in legal demands, and over 2.4 million items removed in 2026. Estimated data based on narrative.

The Growth Story: How Bluesky Scaled to 41 Million Users

Let's start with the growth metrics, because they're fundamental to understanding everything else in this report. Bluesky's user base nearly doubled in 2025. That's not gradual growth. That's exponential.

The company went from 25.9 million users at the beginning of 2025 to 41.2 million by year-end. That's 15.3 million new users in twelve months. To put this in perspective, Twitter added roughly 100 million new users in 2013, but that was from a much smaller base. Bluesky's growth rate is accelerating, not decelerating.

Here's what that growth looks like in terms of content creation: 1.41 billion posts in 2025. That's massive. More impressively, 61% of all posts ever made on Bluesky were created in 2025 alone. This metric reveals something important about the platform's trajectory. It's not just adding passive users who create accounts and disappear. It's attracting engaged users who are actually posting.

Of those 1.41 billion posts, 235 million contained media (images, videos, links). That's about 17% of all posts—which aligns roughly with media attachment rates on other platforms. And notably, those 235 million posts represent 62% of all media posts ever shared on Bluesky. Again, this shows that media sharing is accelerating on the platform.

What's driving this growth? Several factors converge. First, there's the X exodus. When Elon Musk took over Twitter and made controversial changes to moderation, verification, and algorithmic amplification, some users migrated to alternatives. Bluesky benefited significantly from this. Many high-profile journalists, academics, and technologists moved their presence to the platform.

Second, there's the decentralization narrative. After years of criticism about big tech monopolies, the idea of a decentralized social network appeals to users who want more control and less top-down control. Even if most users don't actually run their own servers, the concept is attractive.

Third, there's Bluesky's funding and resources. With backing from major venture capital firms and Dorsey's personal investment, Bluesky can invest in product development, marketing, and safety infrastructure in ways that bootstrapped competitors cannot.

But growth brings complications. More users means more moderation challenges, more legal exposure, and more pressure on infrastructure. The transparency report reveals how severe these pressures have become.

Moderation at Scale: User Reports Jump 54%

Let's talk about the core of Bluesky's moderation operation: the reports that come from users themselves. In 2024, Bluesky received 6.48 million user-submitted moderation reports. In 2025, that number rose to 9.97 million. That's a 54% increase.

Now, here's the important context: Bluesky's user base grew 57% year-over-year. So the increase in moderation reports actually grew slower than the user base growth. In other words, the platform's moderation intensity (reports per active user) actually declined slightly. This suggests that either the user base is becoming better-behaved, or moderation mechanisms are working to discourage frivolous reports.

Breaking down the report categories reveals where the actual problems lie. Here's the distribution:

Misleading content (including spam): 43.73% (4.36 million reports) This was by far the largest category. Within this, spam accounted for 2.49 million reports. The rest falls into general misleading content, likely including false information, clickbait, and misleading headlines.

Harassment: 19.93% (1.99 million reports) This category includes multiple subcategories. Hate speech accounted for about 55,400 reports. Targeted harassment came in at about 42,520. Trolling at 29,500. And doxxing at 3,170. But here's the important detail: most harassment reports fell into a gray area category of "anti-social behavior." This includes rude remarks and general incivility that don't technically violate policy but make the platform unpleasant.

Sexual content: 13.54% (1.52 million reports) Interestingly, the vast majority (1.52 million out of 1.52 million) were about mislabeling—adult content that wasn't properly marked with metadata. Only about 7,520 reports concerned non-consensual intimate imagery. About 6,120 involved abusive sexual content. And over 2,000 involved deepfakes.

Other: 22.14% This catch-all category includes violence, child safety, self-harm, and other violations that don't fit the major categories.

The violence reports (24,670 total) break down into threats or incitement (10,170), glorification of violence (6,630), and extremist content (3,230). These numbers are relatively small compared to harassment and spam, but they're significant because they represent genuine safety concerns.

What's striking about this data is how much it differs from what you might expect. Many people assume that platforms like Bluesky (seen as having lighter moderation) would have massive hate speech and harassment problems. But the data suggests that spam and misleading content are actually the dominant moderation challenges, not hate speech or violence.

Bluesky attributes some of this success to automated systems. The platform's automated moderation tools flagged 2.54 million potential violations in 2025. That's roughly 20% of the 12.51 million total items reported (user reports plus automated flags). This suggests that automated systems are catching problematic content before users report it, reducing the moderation burden on human review teams.

One of the more successful interventions involved reducing the visibility of toxic replies. Bluesky implemented a system that identifies hostile or aggressive replies and puts them behind an extra click, similar to how X handles quote posts and controversial content. This approach reduced daily reports of anti-social behavior by 79%. That's a massive improvement from a relatively simple change.

Month-over-month, Bluesky also saw moderation reports per 1,000 monthly active users decline 50.9% from January to December 2025. This metric is critical because it accounts for user growth. It shows that the platform is becoming moderation-lighter per user as the year progresses. Either users are adjusting their behavior, or the platform is getting better at preventing problematic behavior before it requires reporting.

Bluesky experienced a fivefold increase in government legal requests from 2024 to 2025, rising from 238 to 1,470 requests, indicating significant growth in scrutiny or platform relevance.

The Content Enforcement Numbers: 2.44 Million Items Removed

Beyond reports and flagging, Bluesky actually removed content and accounts from the platform. The numbers are substantial. In 2025, Bluesky took down 2.44 million items, including both accounts and individual pieces of content. Let's break this down.

In 2024, Bluesky removed 66,308 accounts. In 2025, the number rose to 1.02 million accounts removed. That's a 1,443% increase. But wait—the report also mentions that automated systems removed 35,842 accounts in 2024 and 282 in 2025. This discrepancy suggests that the reporting methodology might have changed between years, or that 2024's numbers were incomplete.

For temporary suspensions, Bluesky issued 3,192 in 2025. For permanent removals (ban evasion), the platform issued 14,659. These numbers seem small compared to account removals, but they represent a different category: repeat offenders and users attempting to evade bans.

The majority of account removals focused on inauthentic behavior, spam networks, and impersonation accounts. These are relatively straightforward to identify and remove because they violate clear policy boundaries. An impersonation account pretending to be a celebrity, a spam network coordinating artificially inflated engagement, or a bot farm all have unambiguous removal justifications.

What's more interesting than the removals themselves is the shift in Bluesky's enforcement philosophy. The company labeled 16.49 million pieces of content in 2025, up 200% year-over-year. Meanwhile, account takedowns grew 104%, and individual content removals presumably made up the difference between the 2.44 million total items removed and the 1.02 million accounts removed.

This suggests a deliberate shift toward labeling over removal. Labeling content keeps it visible while adding context. Removal is more aggressive—it eliminates the content entirely. Bluesky appears to be favoring the former approach.

Most labels applied in 2025 involved adult or suggestive content. This makes sense given that the platform has many users who want to control their own content filtering. Instead of removing adult content entirely, Bluesky can label it, and users can choose whether to see it based on their own preferences.

This approach aligns with Bluesky's decentralization philosophy. Different users want different things. Some want maximum protection from adult content. Others want to see everything. Some want slurs labeled. Others don't care. Rather than imposing a single view of what's acceptable, Bluesky is building tools that let users customize their experience.

But here's the catch: this approach only works if the labeling is consistent and accurate. If Bluesky mislabels content, users relying on those labels to filter their feeds will be disappointed or upset. The data from sexual content reports (where 1.52 million of 1.52 million reports were about mislabeling) suggests that labeling accuracy might be a significant challenge.

Influence Operations and Coordinated Inauthentic Behavior

One of the most serious moderation challenges any platform faces is coordinated inauthentic behavior: bot networks, state actors, and organized campaigns designed to manipulate information, spread propaganda, or artificially amplify content.

Bluesky reported removing 3,619 accounts for suspected influence operations in 2025. The company notes that most of these accounts likely operated from Russia. This is significant for several reasons.

First, it shows that authoritarian regimes are already attempting to manipulate Bluesky despite its tiny market share compared to X or TikTok. If they're bothering now, they'll certainly intensify efforts as the platform grows.

Second, Bluesky's ability to identify and remove these accounts suggests that the platform has some level of sophisticated detection capability. Influence operations are hard to spot because the actors behind them employ sophisticated techniques: using VPNs to mask location, creating accounts that appear organic, posting content that seems authentic.

Third, and perhaps most important, this number is likely just the tip of the iceberg. Bluesky caught 3,619 accounts engaged in influence operations. How many did they miss? Comprehensive research on platform manipulation suggests that for every influence operation account caught, multiple others operate undetected. If Bluesky caught 3,619, the actual number operating could be 10,000 or more.

The methodology matters here. Bluesky's detection likely relies on a combination of signals: IP address analysis, account creation patterns, posting behavior analysis, network analysis (who follows whom), and potentially human review. But sophisticated influence operations are designed to evade exactly this kind of detection.

Different countries and actors employ different tactics. Russian operations tend to focus on amplifying divisive content and spreading propaganda about geopolitical events. Chinese operations focus on economic espionage, spreading pro-government narratives, and attacking critics. Iranian and North Korean operations tend to be more varied in their objectives.

The fact that Bluesky's report specifically mentions Russia suggests that Russian operations were the most prevalent or easily identifiable. But it also raises questions: Did Bluesky detect and remove operations from other countries? Are there blind spots in their detection methodology?

For a platform claiming to be better than X when it comes to governance and transparency, influence operations represent an existential threat. If bad actors can coordinate, manipulate, and control narrative on Bluesky just as easily as they do on X, the platform's value proposition evaporates.

Bluesky's user base grew from 25.9 million to 41.2 million in 2025, reflecting a significant increase in user engagement and platform adoption.

The Legal Pressure: Five Times More Requests

Now we get to the number that probably shocked everyone inside Bluesky: government legal requests increased fivefold from 2024 to 2025. In 2024, the platform received 238 legal requests. In 2025, that jumped to 1,470. That's not just an increase—that's explosive growth.

Where are these requests coming from? What are they asking for? The transparency report doesn't provide a detailed breakdown by country or request type, but we can make educated inferences based on what Bluesky has publicly discussed and what typically happens with other platforms.

Likely sources include:

Law enforcement investigating crimes. Police agencies in the US, UK, EU, and elsewhere request user data when investigating crimes like harassment, threats, child exploitation, and fraud. As Bluesky grows, it becomes more valuable to investigators.

Content removal requests. Governments sometimes demand that platforms remove specific content they deem illegal, often content critical of the government, material involving sexual content, or speech that violates local law.

Data requests. Regulators and prosecutors request user data: IP addresses, account creation information, historical posts, contact information, and other details that might be relevant to investigations.

National security requests. Intelligence agencies request information about accounts they believe are connected to terrorism, espionage, or other national security threats.

The fivefold increase raises several questions:

1. Is Bluesky becoming more compliant? Did the platform change its policy on responding to legal requests, making it easier for governments to get data?

2. Are more governments targeting Bluesky? As the platform grows and becomes more visible, do more governments decide it's worth issuing requests?

3. Is Bluesky's data better or more useful? Perhaps requests increased because Bluesky's data is more valuable (more complete, more recent, easier to analyze) than competitors.

4. Are requests concentrated in specific countries? The increase could be driven by one or two countries with aggressive surveillance cultures, or it could be distributed globally.

Without more granular data, it's hard to say. But the transparency report should have included a breakdown by country, request type, and compliance rate. The fact that it didn't—or the fact that we haven't seen that breakdown—is a transparency failure.

For context, let's look at what other platforms report:

Twitter/X reported receiving 3,236 legal requests in the US in the second half of 2024, affecting 6,155 accounts. Extrapolating for full year globally, the numbers are likely in the 5,000-10,000 range globally.

Meta reported receiving 134,000 legal requests in the first half of 2024 globally, a number that has grown consistently year-over-year.

Google received 225,000+ legal requests in the second half of 2023 alone.

Bluesky's 1,470 requests seem low compared to larger platforms, but that's expected given the user base difference. The more important question is the rate of growth. Bluesky's requests grew fivefold. Are other platforms seeing similar acceleration?

One interpretation: Bluesky is becoming mainstream enough that governments are starting to pay attention. Another interpretation: Bluesky's decentralized architecture is making it attractive to bad actors who are then attracting law enforcement attention. A third interpretation: The platform has become a hotbed for activity that various governments want to monitor.

The transparency report doesn't disclose Bluesky's compliance rate with these requests. Did the platform honor all 1,470 requests? Did it challenge some of them? Did it provide data, or did it provide minimal information?

This is crucial because compliance rate reveals whether Bluesky is actually protective of user privacy or just going along with whatever governments demand. A platform that complies with 90% of requests is very different from a platform that challenges half of them on privacy grounds.

Most major platforms publish compliance rates. Twitter typically complies with 80-90% of requests. Meta is similar. But smaller platforms sometimes have lower compliance rates because they have limited legal resources to challenge requests.

Bluesky's failure to publish this data is a red flag. Transparency means providing detailed information, not just headline numbers. If Bluesky wants to position itself as better than X when it comes to privacy and governance, it needs to show that it's actually willing to protect user data against government overreach.

Age Assurance and Youth Safety Measures

One of the most contentious areas in platform moderation today is age assurance: verifying that users are the age they claim to be. This matters because different regulations apply to minors (COPPA in the US, GDPR in the EU) and because platforms have different policies for minors versus adults.

Bluesky's transparency report mentions age assurance compliance, but doesn't provide detailed statistics. This is unfortunate because it's one of the most important areas for platform safety.

Age assurance is hard because:

1. Privacy concerns: Verifying age typically requires identity verification, which creates privacy risks.

2. Inclusivity questions: Identity verification systems often exclude people without official documentation, including homeless people, undocumented immigrants, and people in developing countries.

3. Technical challenges: Age can be verified through payment methods, government ID, biometric analysis, or third-party verification services. Each has limitations.

4. Compliance variation: Different countries have different regulations. What's required in the UK might be prohibited in the EU.

Bluesky's decentralized model adds another layer of complexity. The platform's own instance can require age verification, but independent servers running on the AT Protocol might not. This creates inconsistent protection: a 13-year-old could create an account on Bluesky's main instance and be blocked from adult content. The same child could potentially join an independent instance with no age restrictions.

The transparency report doesn't explain how Bluesky is handling this distributed governance challenge. That's a significant omission because it's core to whether the platform can actually protect minors.

Estimated data suggests that Bluesky's moderation heavily relies on automated flags (50%) and human review (40%), with a smaller portion for automatic removal (10%). This conservative approach minimizes false positives.

Automated Detection: AI-Powered Moderation at Scale

Bluesky's 2.54 million automated flags represent a significant portion of the platform's moderation workload. How does this automated system work? The transparency report doesn't provide technical details, but we can infer from industry practice.

Moderation at scale requires automation because human review is slow and expensive. A single human moderator can review maybe 100-200 items per hour. At that rate, reviewing 12.51 million reported items would require 62,500-125,000 moderator-hours per year. For context, that's equivalent to hiring 30-60 full-time moderators working exclusively on review.

Automatic detection systems use machine learning models trained on examples of policy-violating content. The models analyze text, images, links, and account metadata to assign a violation probability to each piece of content. If the probability exceeds a threshold, the system either removes the content, labels it, reduces its visibility, or flags it for human review.

Bluesky's approach seems to emphasize flagging (sending items to human reviewers) rather than automatic removal. This is more conservative—it avoids false positives where good content gets removed by mistake. But it requires more human labor.

The system's success depends on several factors:

Training data quality. If the training data contains mislabeled examples, the model learns incorrect patterns. If the training data is biased toward certain demographics or types of speech, the model perpetuates that bias.

Model architecture. Different architectures (transformer models, convolutional neural networks, etc.) have different strengths. Models that work great for detecting explicit imagery might fail at detecting subtle hate speech.

Threshold calibration. The system needs to balance false positives (incorrectly flagging good content) against false negatives (failing to catch bad content). Adjust the threshold too high and lots of violations slip through. Too low and you remove acceptable content.

Continuous retraining. Violations evolve. Users develop new coded language to evade filters. Coordinated inauthentic behavior campaigns adopt new tactics. The system needs constant updating.

Bluesky's automated systems caught 2.54 million potential violations out of 12.51 million total reported items. That means about 80% of content reaching the moderation queue came from user reports, and 20% from automation. This suggests that Bluesky's automated detection is helpful but not comprehensive.

For comparison, Meta reports that its automated systems catch 98%+ of hate speech before users report it. Bluesky appears to be lagging in automation capability, or deliberately favoring a more conservative approach that requires human confirmation before removing content.

The Decentralization Question: Can Distributed Moderation Actually Work?

Bluesky's entire value proposition rests on decentralization. In theory, a distributed social network should offer better privacy, fewer outages, greater resilience, and more user control. But decentralization also complicates moderation.

On X, Meta, or TikTok, moderation is centralized. One company makes the rules. One company enforces them. One company can be held accountable.

On Bluesky, moderation is distributed. The AT Protocol specifies the base rules, but individual instances can set stricter policies. Bluesky (the company) moderates its official instance. Other organizations moderate their instances. Users can theoretically choose which instance to join based on moderation preferences.

In theory, this is great. It gives users agency. It prevents any single entity from having monopoly power over speech rules.

In practice, it creates massive inconsistency problems:

1. Divergent policies: Different instances might have different moderation standards. What's allowed on Instance A might be banned on Instance B.

2. Cross-instance content: Users on Instance A can see and interact with content from Instance B, even if Instance B has no moderation standards.

3. Platform reputation: If Instance B becomes a haven for harassment or illegal content, what happens to Bluesky's reputation? Is Bluesky responsible for what happens on third-party instances?

4. Regulatory compliance: Different countries have different laws. The EU requires aggressive removal of hate speech. The US is more hands-off. How does Bluesky comply with both simultaneously on a distributed network?

The transparency report hints at these challenges but doesn't fully address them. Bluesky reported removing 3,619 influence operation accounts. But were all of those on Bluesky's official instance? Or did the company also request that third-party instances remove them?

This distinction matters because it determines whether Bluesky is actually addressing the problem or just cleaning its own house.

Email provides a useful analogy. Email is decentralized. Gmail, Outlook, and Proton Mail are different providers, but they interoperate. However, email also has spam problems that no single provider can fully solve because bad actors operate across provider boundaries. Similarly, Bluesky might struggle to maintain consistent moderation standards because users and bad actors can switch between instances.

The question for Bluesky is whether it can develop governance mechanisms that maintain basic standards across a distributed network without sacrificing the benefits of decentralization. The current approach (Bluesky moderating its own instance, individual instances moderating themselves) seems insufficient.

Spam and misleading content dominate Bluesky's moderation challenges, accounting for 43.73% of reports. Harassment reports are significant but less prevalent, at 19.93%.

Comparisons: How Does Bluesky's Moderation Compare?

Bluesky's moderation report includes lots of numbers, but context is crucial. How do these numbers compare to other platforms?

User base context: Bluesky: 41.2 million users in 2025 X: ~600 million monthly active users Meta/Facebook: ~3 billion monthly active users TikTok: ~1.5 billion monthly active users

Bluesky is roughly 1% the size of X and 0.01% the size of Meta. So you'd expect its moderation numbers to be proportionally smaller.

Moderation reports per user: Bluesky: 9.97 million reports / 41.2 million users = 0.24 reports per user This means roughly 1 in 4 users submitted a moderation report in 2025.

Items removed: Bluesky: 2.44 million items removed As a percentage of content: 2.44 million removed / 1.41 billion posts = 0.17% This means roughly 1 in 6,000 posts was removed.

For comparison, Meta reports removing millions of pieces of content monthly. YouTube removes thousands of videos daily. But per-capita metrics are hard to find because most platforms don't provide them.

Account removals: Bluesky: 1.02 million accounts removed in 2025 As a percentage: 1.02 million / 41.2 million = 2.5% of the user base This is actually quite high. It suggests aggressive enforcement against inauthentic or violating accounts.

Influence operations: Bluesky: 3,619 accounts removed As a percentage of total removals: 3,619 / 1.02 million = 0.35% This is roughly comparable to what other platforms report, though most platforms don't break this out separately.

Overall, Bluesky's moderation appears to be roughly in line with what you'd expect from a platform of its size. It's neither especially aggressive nor especially lenient. The key differentiator isn't the aggressiveness of moderation, but the transparency about it.

Legal and Regulatory Compliance: The Growing Challenge

Beyond user-reported moderation, Bluesky faces complex regulatory compliance challenges. Different countries have different requirements:

The European Union has the Digital Services Act (DSA) and the Online Safety Directive. These require platforms to remove illegal content quickly, provide transparency reports, and protect minors. They also impose liability for algorithmic recommendations and third-party content.

The United States has Section 230 of the Communications Decency Act, which protects platforms from liability for user-generated content. However, it doesn't protect against liability for illegal content (like CSAM or fraud), and new bills are constantly being proposed to restrict or modify Section 230.

The United Kingdom has the Online Safety Bill, which requires platforms to remove illegal content, especially material involving minors.

Russia and China prohibit various types of content and demand that platforms comply with local law or face blocking.

Bluesky's global user base means it needs to comply with all these frameworks simultaneously. That's technically challenging and legally fraught. For example, EU law requires more aggressive removal of hate speech than US law permits. China requires that platforms censor certain topics that US law protects. Bluesky needs to navigate these conflicts.

The fivefold increase in legal requests suggests that regulators are starting to pay attention to Bluesky and demand compliance. As the platform grows, regulatory pressure will only increase.

Data Analysis: What the Numbers Tell Us

Let's synthesize the data into a coherent narrative.

Growth trajectory: Bluesky is growing exponentially. 60% user growth in a single year suggests that the platform has reached escape velocity. It's no longer a niche experiment—it's becoming a real competitor to X.

Content production velocity: 1.41 billion posts in a single year, with 61% of that being new content, shows that Bluesky isn't just acquiring passive users. It's attracting engaged creators who are actively posting.

Moderation intensity declining: User-submitted moderation reports grew 54%, but user base grew 57%. Reports per user actually declined, and month-over-month reports per 1,000 users fell 50.9%. This suggests the platform is becoming cleaner and less problematic per user.

Spam dominates moderation challenges: 43.73% of reports are about misleading/spam content. This is typical of growing platforms—as they scale, spam networks move in to capitalize on the audience.

Harassment is real but manageable: 19.93% of reports concern harassment. This is significant but not overwhelming. It suggests that Bluesky isn't becoming a harassment hellscape, though it's not harassment-free either.

Automated moderation catching 20% of violations: Bluesky's AI systems are flagging about 2.54 million items, roughly 20% of the moderation queue. This is lower than industry leaders like Meta, suggesting either less sophisticated AI or a deliberate choice to rely more on human judgment.

Enforcement shifting toward labeling: Bluesky applied 16.49 million labels (200% increase) while removing content more conservatively. This aligns with the platform's decentralization philosophy: let users choose what they see rather than making top-down decisions.

Influence operations are present but manageable: 3,619 removed accounts represent a small percentage of total removals, but Russia-linked operations were identified and addressed.

Government pressure escalating rapidly: The fivefold increase in legal requests is the most dramatic number in the entire report. This suggests that Bluesky is becoming significant enough that governments want access to user data.

Transparency Report Gaps: What Bluesky Didn't Tell Us

The transparency report is more comprehensive than most platforms' early reports. But it still has significant gaps.

No country-by-country breakdown of legal requests: Which countries issued the 1,470 requests? Did they come from authoritarian governments, democratic countries, or a mix? Bluesky should publish this.

No compliance rate for legal requests: Did Bluesky comply with all 1,470 requests, or did it challenge some on privacy grounds? This is crucial for assessing whether the platform actually protects user data.

Incomplete influence operation detail: Bluesky mentioned Russia-linked operations but didn't disclose whether it found operations from other countries. Are there blind spots in the detection?

No granular moderation appeal data: When Bluesky removes content or suspends accounts, users can appeal. How many appeals succeeded? This would show whether the moderation system has error rates.

Limited sexual content detail: Bluesky reported that 1.52 million sexual content reports were about mislabeling. But how many of those labels were correctly fixed? If most were ignored, that's a problem.

Insufficient age verification data: Bluesky mentions age assurance compliance but provides no statistics on how many minors are on the platform, what protections they receive, or whether third-party instances follow the same rules.

No comparison to 2024 moderation: Bluesky said it previously issued moderation reports in 2023 and 2024, but doesn't publish links to those reports or explain how 2025 numbers compare to previous years (except for the legal requests section).

Limited algorithmic recommendation data: Bluesky's report doesn't detail how its recommendation algorithm works or whether it's tested for bias or harms.

A truly transparent platform would address all of these gaps. Bluesky's report is a good start, but incomplete.

What This Means for Bluesky's Future

The transparency report reveals a platform that is growing rapidly, moderating aggressively, but facing mounting pressure from governments and increasing complexity as it scales.

Growth momentum is real: 60% annual growth suggests that Bluesky isn't a fad. Users are genuinely migrating to the platform.

Moderation at scale is hard: Even with 16.49 million labels and 2.44 million removals, bad content still exists on the platform. The 54% increase in user reports suggests that users are continuing to find violations.

Decentralization adds complexity: Bluesky needs to develop governance mechanisms that work across a distributed network. The current approach seems insufficient.

Government pressure is mounting: The fivefold increase in legal requests is just the beginning. As Bluesky grows, regulatory pressure will intensify.

Transparency is becoming table stakes: Bluesky's release of this report signals that transparency is now expected from social platforms. The question is whether the transparency is genuine or just theater.

For users considering whether to move to Bluesky, the report suggests a platform that is trying to do moderation right, but still learning. It's not perfect. But it's also not worse than X or Meta from what the data shows.

Industry Implications: What Other Platforms Can Learn

Bluesky's transparency report is instructive for the entire platform ecosystem.

Transparency builds trust: The simple act of publishing detailed moderation statistics helps users understand how the platform operates and whether it's trustworthy.

Distributed moderation is possible but hard: Bluesky's decentralized model offers benefits, but it also complicates moderation. Other platforms exploring federation should learn from Bluesky's experience.

Spam and misleading content are the real problem: Most conversations about platform moderation focus on hate speech and harassment. But Bluesky's data shows that spam and misleading content are more prevalent. Platforms need to invest in solutions for these problems.

Government pressure is inevitable: As platforms grow, governments will demand data access. Platforms need policies and legal teams prepared to respond to these requests while protecting user privacy.

Labeling is often better than removal: Bluesky's shift toward labeling content rather than removing it offers a middle ground between censorship and anything-goes policies. Other platforms should experiment with similar approaches.

The Bigger Picture: Social Media in 2026

Bluesky's transparency report comes at a fascinating moment in social media history. X is under pressure from advertisers and regulators. Meta is facing scrutiny over algorithmic harms. TikTok faces potential bans. Threads is struggling to find product-market fit.

Into this fragmented landscape, Bluesky is growing. It's not because Bluesky is perfect. It's because users are frustrated with the existing alternatives and willing to try something new.

The platform's success depends on whether it can maintain trust as it scales. Trust requires transparency, consistent moderation, and accountability. Bluesky's transparency report is a step in that direction.

But the report also reveals challenges that the platform will struggle with for years: how to moderate consistently across a distributed network, how to respond to government demands while protecting user privacy, and how to prevent bad actors from exploiting the platform's growth.

These aren't unique to Bluesky. Every social platform faces them. What's different is that Bluesky is attempting to address them transparently, from the beginning, rather than waiting until the platform becomes a scandal.

That might not be enough to guarantee Bluesky's long-term success. But it's a better foundation than most platforms started with.

Recommendations for Bluesky

If Bluesky is genuinely committed to being better than existing platforms, it should:

1. Publish country-by-country legal request data: Transparency demands this. Users deserve to know which governments are demanding their data and how Bluesky responds.

2. Implement consistent moderation across instances: Develop standards that independent instances should follow, and provide tools to help them comply.

3. Invest in influence operation detection: 3,619 removed accounts suggests that Russian operations are present. Bluesky should invest heavily in detecting and disrupting these campaigns.

4. Publish moderation appeal success rates: Show how many content removals and account suspensions are overturned on appeal. This demonstrates whether the moderation system is fair.

5. Develop better age verification: Implement age assurance that actually works without compromising privacy or excluding vulnerable users.

6. Improve automated detection accuracy: Bluesky's 20% automation rate suggests room for improvement. Invest in AI systems that can catch more violations without increasing false positives.

7. Explain algorithmic recommendation: Publish details about how the algorithm works, what signals it uses, and whether it's tested for bias or manipulation.

8. Implement user appeals dashboards: Let users see why content was removed and provide clear paths to appeal decisions.

FAQ

What is Bluesky's transparency report?

Bluesky's first comprehensive transparency report, published in January 2026, documents the platform's moderation actions, legal compliance, account removals, and other safety metrics throughout 2025. It includes statistics on user-submitted reports, automated detection, content labeling, account suspensions, and government legal requests. The report covers an exceptional year of growth: Bluesky doubled its user base from 25.9 million to 41.2 million users, created 1.41 billion posts, and faced a fivefold increase in legal demands from law enforcement and government agencies.

Why did Bluesky issue a transparency report?

Bluesky released the report to demonstrate accountability, build user trust, and show how the platform approaches content moderation and safety challenges. As a platform explicitly positioning itself as an alternative to X and other established social networks, Bluesky needed to prove that it takes moderation seriously and operates responsibly at scale. Transparency reports have become expected documentation for major social platforms, as they demonstrate that companies are managing content, responding to legal requests, and maintaining platform health. Bluesky's timing is significant because the platform experienced exceptional growth in 2025, requiring stronger evidence that scaling hasn't compromised safety.

What were the key statistics in Bluesky's transparency report?

The report included several major findings: Bluesky grew 60% to reach 41.2 million users; users generated 1.41 billion posts (61% of all posts in platform history); 9.97 million user-submitted moderation reports were filed; Bluesky applied 16.49 million content labels; the platform removed 2.44 million items including 1.02 million accounts; 3,619 accounts were removed for suspected influence operations (mostly Russian); and crucially, government legal requests increased fivefold from 238 in 2024 to 1,470 in 2025. The largest category of reports involved misleading content and spam (43.73%), followed by harassment (19.93%) and sexual content (13.54%).

How does Bluesky's moderation compare to other platforms?

Bluesky's moderation intensity appears broadly comparable to other platforms when adjusted for user base size. The platform removes roughly 0.17% of posts and 2.5% of accounts annually, which is in line with industry norms. However, Bluesky's approach differs in philosophy: it favors labeling content over removing it, allowing users to customize their moderation experience. This aligns with Bluesky's decentralization philosophy. Unlike centralized platforms like Meta or X, Bluesky's distributed architecture means that moderation varies across independent instances running on the AT Protocol. Bluesky's 2.54 million automated flags represent about 20% of the total moderation queue, suggesting less sophisticated AI automation compared to industry leaders like Meta, which reports catching 98%+ of hate speech through automated systems.

What does the fivefold increase in legal requests mean?

Government legal requests jumped from 238 in 2024 to 1,470 in 2025, indicating that law enforcement, regulators, and prosecutors are increasingly targeting Bluesky. This reflects the platform's growing prominence and user base. However, the transparency report doesn't disclose which countries issued requests, what they requested, or how many Bluesky complied with. This is a significant transparency gap because it obscures whether authoritarian governments are demanding data for surveillance purposes or whether requests are primarily from democratic countries investigating legitimate crimes. The increase also suggests that Bluesky is becoming valuable enough for governments to monitor and demand access to user information, a concern for privacy-conscious users who chose the platform partly for its decentralization promises.

What are the main moderation challenges Bluesky faces?

Bluesky's report reveals that spam and misleading content dominate moderation workloads (43.73% of reports), followed by harassment (19.93%) and sexual content (13.54%). The platform also struggles with mislabeling of adult content: 1.52 million of 1.52 million sexual content reports concerned adult material that wasn't properly tagged. Additionally, Bluesky faces challenges with coordinated inauthentic behavior (3,619 influence operation accounts removed, mostly Russian-linked), toxicity management (79% reduction in anti-social behavior reports after implementing a visibility-reduction system), and the inherent complexity of moderating consistently across a decentralized network where independent instances maintain different standards. The month-over-month decline in reports per 1,000 users (down 50.9%) suggests the platform is improving, but the presence of nearly 10 million reports in a single year indicates that moderation challenges remain significant.

How does Bluesky's decentralized architecture affect moderation?

Bluesky's decentralized architecture, built on the AT Protocol, fundamentally changes how moderation works compared to centralized platforms. Instead of one company (Bluesky) controlling all moderation decisions, users can join different instances (servers) operated by different organizations with different moderation policies. Bluesky can moderate its own instance, but it can't unilaterally control content on independent instances. This creates governance challenges: consistent moderation standards become difficult to maintain, cross-instance content visibility creates exposure to material moderated elsewhere, and regulatory compliance becomes complicated when different instances follow different rules. However, the architecture also offers benefits: users gain more control over their moderation experience, content can be labeled rather than removed, and no single entity monopolizes speech control. The success of decentralized moderation ultimately depends on whether communities can develop shared governance mechanisms while respecting the autonomy that decentralization promises.

What should users know about Bluesky's privacy and legal compliance?

The fivefold increase in government legal requests should make users aware that Bluesky's data is increasingly valuable to law enforcement and governments. The report doesn't disclose Bluesky's compliance rate, which is a red flag: users deserve to know whether the platform protects their data against government overreach or routinely complies with requests. For users who chose Bluesky partly for privacy reasons, this lack of transparency is concerning. Additionally, Bluesky's decentralized architecture means that privacy protections can vary across instances. Independent servers might have weaker privacy practices than Bluesky's official instance. Users should be cautious about assuming that Bluesky offers superior privacy compared to mainstream platforms without seeing more detailed privacy policies and compliance data. The platform shows promise but needs more granular transparency about how user data is handled when governments request it.

What does Bluesky's success mean for the future of social media?

Bluesky's 60% growth and 41.2 million user base suggest that users are willing to migrate to alternatives when they become sufficiently frustrated with existing platforms. This creates space for new competitors and puts pressure on X, Meta, and other established platforms to address user concerns about content moderation, privacy, and algorithmic control. If Bluesky can maintain trust through transparency and consistent moderation as it scales, it could establish itself as a genuine alternative to X. However, the transparency report also reveals that Bluesky faces the same moderation challenges as larger platforms: spam, harassment, influence operations, and government pressure. This suggests that no social platform can escape these challenges through design choices alone. The industry will likely continue fragmenting, with different platforms attracting users with different moderation philosophies and governance approaches. Bluesky's transparency report sets a precedent that other platforms may need to match.

Conclusion: A Platform Learning in Public

Bluesky's first comprehensive transparency report is significant not because the numbers are shocking or unexpected, but because the platform is learning to operate with accountability from the beginning rather than decades in.

The data tells a story of explosive growth, robust moderation efforts, and mounting regulatory pressure. Nine-point-nine-seven million user reports show that the community is engaged in safety. Two-point-four-four million items removed demonstrate that enforcement is occurring. And the fivefold increase in legal requests signals that governments are starting to pay attention.

What's remarkable is the decentralized context. Bluesky achieves these moderation results without centralized control, without a recommendation algorithm optimized for engagement, and without the surveillance infrastructure that makes Meta and Google powerful. This matters because it suggests that alternative models are technically feasible.

But the report also reveals challenges that Bluesky will struggle with for years: maintaining consistent moderation across a distributed network, protecting user privacy against government demands, and preventing bad actors from exploiting the platform's growth. These aren't unique to Bluesky. Every social platform faces them.

The question isn't whether Bluesky can solve these problems perfectly. No platform can. The question is whether Bluesky maintains the transparency and accountability that built trust in the first place as the platform scales and inevitably makes controversial moderation decisions.

The first report is promising. But transparency is a commitment, not a moment. Bluesky needs to maintain and deepen this transparency in subsequent reports. Only then can users truly evaluate whether the platform is different from what came before.

Key Takeaways

• Bluesky grew 60% in 2025 from 25.9M to 41.2M users, generating 1.41B posts with 61% created in the final year
• User-submitted moderation reports reached 9.97M (up 54%) covering spam/misleading (43.73%), harassment (19.93%), sexual content (13.54%)
• Government legal requests exploded fivefold from 238 in 2024 to 1,470 in 2025, suggesting increasing regulatory scrutiny
• Bluesky shifted enforcement strategy toward content labeling (16.49M labels, +200%) versus removal (2.44M items), enabling user-controlled moderation
• Decentralized architecture creates unprecedented moderation complexity requiring consistency across independent instances with varying policies

Related Articles

• Meta Blocks ICE List: Content Moderation, Privacy & Free Speech [2025]
• Meta Blocks ICE List: Content Moderation or Political Censorship? [2025]
• TikTok US Ban: 3 Privacy-First Apps Replacing TikTok [2025]
• Gmail Spam & Filtering Issues: Google's New Fix Explained [2025]
• TikTok's New Owners & What It Means for Your Feed [2025]
• Age Verification & Social Media: TikTok's Privacy Trade-Off [2025]

FAQ

What is Bluesky's First Transparency Report: A Deep Dive Into Moderation, Legal Demands, and Growth [2025]?

For years, social media companies operated in the shadows when it came to moderation decisions and legal compliance

What does introduction: the moment social media platforms started showing their work mean?

They'd announce policy changes, remove accounts, and hand over user data without much fanfare or explanation

Why is Bluesky's First Transparency Report: A Deep Dive Into Moderation, Legal Demands, and Growth [2025] important in 2025?

Platforms realized that transparency reports weren't just nice to have—they were essential to building trust, demonstrating accountability, and proving they weren't completely out of control

How can I get started with Bluesky's First Transparency Report: A Deep Dive Into Moderation, Legal Demands, and Growth [2025]?

Bluesky, the decentralized social network founded by Jack Dorsey as an alternative to X (formerly Twitter), just took a major step in that direction

What are the key benefits of Bluesky's First Transparency Report: A Deep Dive Into Moderation, Legal Demands, and Growth [2025]?

In January 2026, the company released its first comprehensive transparency report, and the data tells a fascinating story about a platform scaling rapidly while wrestling with the same moderation challenges that plague every major social network

What challenges should I expect?

Here's what makes this moment significant: Bluesky isn't just another social media competitor