# Privacy at a cost: the dark web’s main browser helps pedophile networks flourish **URL:** https://discuss.privacyguides.net/t/privacy-at-a-cost-the-dark-web-s-main-browser-helps-pedophile-networks-flourish/30501 **Category:** News **Tags:** article **Created:** 2025-08-27T05:16:51Z **Posts:** 36 ## Post 1 by @rlogin — 2025-08-27T05:16:51Z > **[Privacy at a cost: the dark web’s main browser helps pedophile networks...](https://www.theguardian.com/technology/2025/aug/25/tor-network-child-sexual-abuse)** > > The Tor network’s privacy architecture creates a safe haven for predators to share child sexual abuse material Additional info on Reddit: [Reddit - The heart of the internet](https://www.reddit.com/r/TOR/comments/1mzywgy/a_new_attack_on_tor_paid_for_by_the_canadian/) --- ## Post 2 by @rlogin — 2025-08-27T03:00:11Z It is astonishing and even comical how they portray Tor and the Tor Project. They are making it seem like it’s a centralized app, like Telegram or something. Look: > **The platform’s** design makes it virtually impossible to remove harmful posts or illegal content, they say, and the organization behind Tor has resisted pressure to implement even basic safeguards. > The Canadian Centre for Child Protection (C3P) told the Guardian it had submitted more than 19,000 notifications to the Tor Project – the privacy-focused US non-profit that develops and maintains the network – flagging child sexual abuse material (CSAM) **detected on its system**. Yet the organization has taken no action, according to C3P. > “Tor is designed in a way such that they can’t remove anything **and they refuse to change it,** because the idea of any sort of censorship on there is bad,” said Lloyd Richardson, C3P’s director of technology. “Tor has a board of directors that makes decisions surrounding this, and **I don’t think they’ve ever been held to account for any of this**. They’re the only people who can essentially intervene, and they refuse to do so.” > Neither of Tor’s products has systems to detect, moderate or report CSAM, **or mechanisms for processing user-generated content in the way mainstream social networks do.** > The experts interviewed urged Tor to act to remove CSAM **from its sites** and stem the formation of pedophile groups. --- ## Post 3 by @beantaco — 2025-08-27T05:04:13Z Now The Guardian is spreading the “protect the children” narrative that Western governments are peddling. They may claim to have written a balanced piece, and declared they operate [their own onion service](https://www.guardian2zotagl6tmjucg3lrhxdk4dw3lhbqnkvvkywawy3oqfoprid.onion/), but (via the headline, headings, focus, positioning of content, etc.) it overall brings criticism and stigma the Tor Project and the Tor network more than it supports Tor’s goals and benefits. Its main aim was to bring attention to illicit Tor use for CSAM purposes. > [@rlogin](#): > > It is astonishing and even comical how they portray Tor and the Tor Project. They are making it seem like it’s a centralized app, like Telegram or something. I agree the article’s wording is poor, however, it looks like some of the poor wording comes straight from Tor’s critics. The article also says > Tor is not considered an [electronic service provider] since it is engineered for privacy and anonymity, routing internet traffic to conceal user identities rather than hosting or moderating content. To me the important points for the privacy community, us, are 1. We should assume Tor is or will very soon be in the crosshairs of the “protect the children” agenda, like VPNs currently are in the UK. 2. I believe we should acknowledge and discuss the problem of illicit use of privacy technologies, not ignore nor downplay it, while standing firm that privacy is a human right and crackdowns ought to be on illicit activities themselves and not technologies we use and depend upon. --- ## Post 4 by @jonah — 2025-08-27T05:13:35Z It’s as absurd as saying DNS is responsible for the distribution of CSAM, but unfortunately in Europe that exact line of thinking has been successfully used in courts by media companies against torrent sites. > The experts interviewed I guess “experts” means “some randos with an opinion” to _The Guardian_. Very useful info to know when reading their content! --- ## Post 5 by @_TrustyRocinante — 2025-08-27T05:42:44Z No privacy for anyone = the children are finally safe Privacy for anyone = everyone is a pedo But seriously, yes this is how privacy (or rather, anonymity), works. You give up some control over your people so that they have the right to privacy. This means bad people can have privacy too, which obviously is not good for CSAM. But what’s the alternative? Well, government has complete control over it’s people. Every citizen’s digial message, conversation, photo, and written thought is accessed by the government so that they can potentially find some baddies. These agencies wrongly belive the only way to 100% eradicate CSAM is for every single person in the world to give over every shed of privacy to the government. There will always be another aspect of privacy to give up for “the children’s safety”. What’s next? Cameras in our homes to stop child abuse? ID to connect to any Wi-Fi network? It sounds a bit farfetched. But no government that is wanting to truly eradicate CSAM will be satisfied if Tor just vanished. There will be something else to lean on. --- ## Post 6 by @jonah — 2025-08-27T05:48:38Z > [@_TrustyRocinante](#): > > The only way to 100% eradicate CSAM is for every single person in the world to give over every shed of privacy to the government. I don’t want people to be fooled into thinking this could even theoretically work. Even if we give up all privacy, there will still be CSAM and other problems in the world. These problems are not caused by having privacy. --- ## Post 7 by @_TrustyRocinante — 2025-08-27T05:57:17Z Yep, totally agree. Edited it to clarify that it’s a bogus theory, but a theory that most governments hold. I have sympathy for the activists or agencies with the goal to protect children against CSAM, but I don’t think throwing privacy out the window is even a remotely good solution. --- ## Post 8 by @mangomango — 2025-08-27T07:04:00Z > [@jonah](#): > > These problems are not caused by having privacy. Instead of privacy causing crime, I think that the argument is that privacy allows criminals to cover their tracks and not be caught ? --- ## Post 9 by @fria — 2025-08-27T07:08:14Z > [@jonah](#): > > It’s as absurd as saying DNS is responsible for the distribution of CSAM, I hear they’re using something called “internet protocol” to send CSAM these days, maybe the government should ban that :thinking: --- ## Post 10 by @shadowwwind — 2025-08-27T08:17:52Z IETF has never been held accountable --- ## Post 11 by @Encounter5729 — 2025-08-27T15:41:09Z Okay, so here’s my opinion: we shouldn’t dismiss those concerns. However, let’s think about solutions and what they would do in practice. 1. Ban Tor entry nodes Everyone who wants to use TOR will now use a VPN, making the fact they use Tor hidden. Congrats, know law enforcement are even less capable of identitying those awful criminals. 1. Ban Tor Organisation from developing Tor: Development likely to continue underground, reduce accessibility of Tor. The later will mostly hurt legit users of Tor. 1. Mandate Tor to ban domains Not sure here, someone knowledgeable could pitch in whether this is possible and how. In summary, I think we shouldn’t dismiss those concerns but let’s have constructive discussions, not hitpieces like this article. Often, Privacy get suppressed to counter evil actor X. Let’s not be in the trap of binary choices either, we can do better. --- ## Post 12 by @anonymous378 — 2025-08-27T15:50:08Z This just sucks for the actual issue of CSAM as it is constantly used as a bad faith reason against legitimate tools and takes aways real efforts to curb the issue. That’s all without even thinking about the harms to privacy and other rights these bad faith efforts have. --- ## Post 13 by @jonah — 2025-08-29T02:53:05Z The reality is that people who are actually involved in internet safety and privacy advocacy have been thinking about this problem for _decades_, and the consensus is that these proposals are untenable. The only people who disagree fall into two camps: 1. People who have _not_ thought about these problems for decades, but rather read about them and formulated a gut opinion in the last 5 minutes; and 2. Lobbyists and politicians who have ulterior motives, like ultra-religious think tanks who know that once the regulations are in place it will be much easier to brand _anything_ they don’t like as pedophilic. These concerns are dismissed because the opposing viewpoint is not a logical opinion, it’s a belief no less absurd than believing the world is flat. The only way you could believe it is if you create the belief in your mind and then do not follow it to its logical conclusion or seek out any evidence that contradicts your worldview. You immediately came up with some solutions, and then immediately followed them up with reasons they would not work. That is the line of logical thinking that I think we should encourage people in the first camp to also do, and that would be more helpful than entertaining a belief system that the only way crime can be solved is with an omnipresent government :slight_smile: --- ## Post 14 by @anon92357554 — 2025-08-29T03:11:44Z I am not sure what the solution is. I don’t think messing with Tor is it. At the same time, I was pretty startled to read that 1 out of every 10 searches on Tor is for CSAM --- ## Post 16 by @jonah — 2025-08-29T05:07:46Z > [@anon92357554](#): > > 1 out of every 10 searches on Tor is for CSAM Not on Tor. On Tor-specific search engines. Remember the default search engine in Tor Browser is DuckDuckGo, and Tor is largely used to browse clearnet sites. Some stats place hidden service traffic as low as 3%-8% of total Tor traffic, and you are talking about _at most_ 1 in 10 of that.[[1]](#footnote-108861-1) * * * 1. Someone will read this and think “well, why don’t we just scrap hidden services entirely then?” and if that is you reading this then you should reply so I can argue against that point too lol [↩︎](#footnote-ref-108861-1) --- ## Post 17 by @ImTooPhaT — 2025-08-29T14:22:09Z According to The Guardian _**“A study published by [Nature](https://www.nature.com/articles/s41598-024-58346-7)** last year determined that CSAM is easily available using 21 out of the 26 most-used Tor search engines, and that at least 11% of users’ searches seek CSAM.”_ The Nature article itself contains paragraphs about overcoming CSAM addiction \*“When we study the searches, we discover that there are a few hundred queries from people who want to cease viewing CSAM…” \*Even if it’s just a small fraction compared to the number of queries from people who are looking for CSAM, Tor would still be the best online platform for these people. I have never typed “how do I overcome my attraction to children” on Google and I don’t want to find out what would happen if I do, but I will assume the worst. Best case scenario, I would find resources or therapists, but using Tor is the best way of avoiding false positives. As for the “solution”, the Guardian article talks about Tor as a network for CSAM distribution, but it doesn’t talk about how CSAM material was created in the first place for distribution. _“Research published in the Journal of Online Trust and Safety indicates [that more than 40%](https://tsjournal.org/index.php/jots/article/view/29/17)of dark web users accessing CSAM go on to **seek contact** with children with the aim of pursuing sexual acts.”_ How are they contacting these children? Through Tor? That seems extremely unlikely. They will most likely try to contact minors on online platforms with a large user base of children, pretends to be a child and groom them into creating CSAM material. These platforms supposedly have _“systems to detect, moderate or report CSAM, or mechanisms for processing user-generated content”_ (something that’s lacking on Tor), but when these platforms are in the business of maximizing returns for their investors, their principles on user safety becomes less of a priority (if not abandoned). I would say that’s worse than not having these systems in place. If the Guardian is serious about child safety, they (and other publishers) should push for more accountability from these online platforms that are facilitating the abuse of children, and the general public should cultivate around this issue as well instead of allowing their governments to slap a nonsensical regulation like the UK Online Safety Act. Of course, this is easier said than done. I hear Roblox is a haven for child predators too. --- ## Post 18 by @anon92357554 — 2025-08-29T14:59:08Z Good correction. Overall, it is a catch-22 in that it is a smaller number of people relative to the rest of the internet, yet as a ratio of user searches, my (uneducated) guess is that 10% is +10x the rates on Google et al. Hypothetically, if someone waved a magic wand and Tor had Chrome’s market share overnight, would we be OK with 10% of tor search-engine traffic looking up CSAM? What about 20%? 50%? …I can see the slippery slope that leads to requiring IDs/banning VPNs and I don’t like it, but I too can see the slow emboldening of ‘niche’ interests as well when they can grow unchecked in the recesses of the internet. Is there a threshold where the cost of privacy is too great? I do not have an answer, but it does remind me of the conversation around censorship in the US over the last decade… Whenever someone tries to deplatform hate speech or obvious misinformation (ie: RT News), there are people decry it as 1984/Big Brother. They have a clear example of a worse case scenario… OTOH, the ‘worst case scenario’ for unchecked misinformation/hate speech doesn’t have a clear analogy that we can point to. So instead, there is just a slow erosion of norms via a steady increase in an increase in hate crimes, an unsurety of what is real, that is further complicated by growing deep fakes/AI slop, a mistrust in expert opinions around science/government/health etc. --- ## Post 19 by @jonah — 2025-08-29T14:59:56Z > [@ImTooPhaT](#): > > How are they contacting these children? Through Tor? That seems extremely unlikely. No, [**[edit](https://discuss.privacyguides.net/t/privacy-at-a-cost-the-dark-web-s-main-browser-helps-pedophile-networks-flourish/30501/21)**]: if this happens on Tor, then it’s likely they are contacting other adults on Tor who have children, not children directly. > [@ImTooPhaT](#): > > they (and other publishers) should push for more accountability from these online platforms My question for you is: how would this be implemented differently than the Online Safety Act? --- ## Post 20 by @ImTooPhaT — 2025-08-29T16:42:02Z > [@jonah](#): > > No, it’s likely they contact other adults on Tor who have children, unfortunately. Maybe I’m blissfully ignorant, but how does this lead to grooming compare to my Roblox example? Are the parents allowing their child to talk to other adults through Tor? Or is this more about trafficking? > [@jonah](#): > > My question for you is: how would this be implemented differently than the Online Safety Act? So my Roblox example: they don’t have to do anything differently. They just have to demonstrate how they value the safety of their users with the safeguarding tools they already have. Do they respond to reports over growing concern of child predators? --- ## Post 21 by @jonah — 2025-08-29T17:08:18Z > [@ImTooPhaT](#): > > Or is this more about trafficking? That would be my assumption. > _Research published in the Journal of Online Trust and Safety indicates [that more than 40%](https://tsjournal.org/index.php/jots/article/view/29/17)of dark web users accessing CSAM go on to **seek contact** with children_ I see now that I think you are interpreting “contact” as communication, but I think the study is actually referring to direct physical contact. Edit: What the study actually says is that 42% of respondents who viewed CSAM on the dark web then went on to seek “direct contact with children through online platforms” afterward. Online platforms in this case is not necessarily Tor, to your point it could very well be Roblox, etc. Maybe you know that and I misunderstood what you were originally asking this whole time lol. All I meant to say that **if** this is happening on Tor then it is likely between adults arranging this trafficking, but I agree it certainly is more likely to just be happening on children’s games and social media in many cases. --- ## Post 23 by @Laitinlok — 2025-08-29T20:00:56Z They should ban https to protect the children /s --- ## Post 24 by @ImTooPhaT — 2025-08-29T22:54:49Z Well, yeah, that’s common knowledge, but we’re talking about meeting people online. --- ## Post 25 by @ImTooPhaT — 2025-08-29T23:02:52Z > [@jonah](#): > > That would be my assumption. That would be different. I’m not sure about trafficking, but the Guardian article was more about distribution of CSAM material. > [@jonah](#): > > “direct contact with children through online platforms” Ah, the Guardian article worded it a bit differently, but I did guess it would be a popular online platform like Roblox. --- ## Post 26 by @mangomango — 2025-08-30T20:44:46Z > [@jonah](#): > > My question for you is: how would this be implemented differently than the Online Safety Act? Defaulting the accounts to private maybe? Restricting who can DM you by default? Explaining the dangers to children (when setup, when DM, tutos)? Disallowing phone phone number discovery for people who say they are under 18? Encouraging parents (and users in general)to not post pictures about their children (or people who do not consent in general)? Enforcing e2e DMs worldwide? Not encouraging users as much to post pictures ? Not asking for the real names of users, especially warning about using their real name in the username? These are just some ideas :person_shrugging: --- ## Post 27 by @SwampTrainer — 2025-09-01T06:49:23Z I would bet money they’re experts in opinion-holding, which is why they’re not named. Some windbag that had a headshot and not a resume. --- ## Post 28 by @whattheheehaw — 2025-09-01T09:23:09Z Well, the author not only confindent enough to have a strong opinion about a technology she doesn’t understand — an opinion which uses fear and ignorance to give the oligarchy total control over communication — speaks to something dark. We are so far removed from true tyranny in the West that we, apparenly, are going to have to suffer through it again (and over come it again and again and again). It’s the Anti-Enlightenment, boys and girls! The cycle begins again, I’m afraid. Sadly, I’m not old so I’m going to have to suffer through it. --- ## Post 29 by @flossposse — 2025-09-02T04:15:57Z Disappointing to see this yellow journalism from The Guardian. I’m going to highlight a few of the ways this piece misrepresents research or is otherwise misleading: Katie McQue, the article’s author, refers to “A study published by Nature” and “Research published in the Journal of Online Trust and Safety”. This makes it sound like two independent studies, but in reality it’s the same research project, conducted by the same people, in both journals. The study published in 2022 in the Journal of Online Trust and Safety is based on a subset of the data used in the study published in Nature in 2024. The study’s lead author is Juha Nurmi, the operator of [ahmia.fi](http://ahmia.fi). Ahmia is a special-purpose search engine - it only indexes, and returns results for, onion sites. The study’s data on search queries is drawn entirely from searches performed on Ahmia. As such, this data is not representative of most tor usage. The study authors specifically acknowledge this fact, stating, “Tor is mainly used to access legal clear web content, as only 6.7% of Tor users access hidden onion services (Jardine, Lindner, and Owenson 2020)”. When McQue says “at least 11% of users’ searches seek CSAM” she’s creating the impression that 11% of tor users are searching for CSAM, but in reality what the study indicates is that 11% of search sessions _on Ahmia_ include CSAM terms. The searches on Ahmia represent at most a single-digit percentage of all tor searches, and given their exclusive focus on onion sites, the portion of their searches aimed at CSAM is certain to be vastly higher than searches performed elsewhere using tor. The study authors do not attempt to estimate how many unique individuals, and how many bots, are responsible for the traffic on Ahmia. Later in the article, McQue quotes Nurmi as a “cybersecurity expert and postdoctoral research fellow at Tampere University in Finland”, but makes no mention of the fact that he’s the operator of Ahmia and the author of the two studies she referenced earlier. McQue uses the header “Millions of pedophiles”, apparently based on Europol reporting that Kidflix had 1.8 million users and another site’s claims of having a million users. But the anonymous nature of tor makes it very difficult to tell whether two different page loads, or even two different accounts on a darknet site belong to two different individuals, or whether they are in fact the same person. If it were possible to accurately distinguish between unique visitors using tor, then tor wouldn’t be blocked by cloudflare on half the web like it is today. Good journalism would give readers an accurate understanding of the statistics. Good journalism would make the reader aware of all the caveats and limitations in the methodologies. Good journalism would clarify what tor is, what “dark web” means, and how those are different. It would make clear that tor is a transport, not a hosting platform, and would help the reader understand that the statement, “experts interviewed urged Tor to act to remove CSAM from its sites” makes about as much sense as “experts interviewed urged the highway to act to remove illegal substances from shops and storefronts” Ending child abuse is a worthy cause. Using the issue as a bogeyman to undermine civil liberties is not. McQue’s next article might well be titled: “Security at a cost: door locks help human traffickers imprison victims”. --- ## Post 30 by @SwampTrainer — 2025-09-02T05:01:58Z Wow. This is straight up propaganda, then. Not surprising, but still. --- ## Post 32 by @Robin_RM — 2025-09-09T16:23:21Z > [@ImTooPhaT](#): > > How are they contacting these children? Through Tor? That seems extremely unlikely. In the anonymous survey, they only asked the following question: «_How often after viewing CSAM/illegal violent material have you sought direct contact with chlldren online?_» ([p. 11](https://tsjournal.org/index.php/jots/article/view/29/17)), They did not specify the methods, so it could have been outside Tor, or they could have lied. In that same report, they acknowledge this limitation: «_Seto, Hanson, and Babchishin (2011) also point out that some self-reported offenses may be false confessions._» (p. 10). She later cites that these findings are consistent with other studies that correlate CSAM consumption with child sexual abuse: «_This is similar to Seto, Hanson, and Babchishin (2011)'s meta-analysis, which found that 55% of CSAM offenders admitted to contact sexual offense in self-report studies._» (p. 17), However, these studies do not control for the order in which events occur. The same author, Babchishin, later conducted a [study](https://journals.sagepub.com/doi/10.1177/00938548211040849) finding that it is not common for CSAM offenses to be the first in a criminal career; in many cases, offenders actually escalate from child sexual abuse to CSAM consumption. This would be consistent with a recent [meta-analysis](https://www.sciencedirect.com/science/article/pii/S1359178925000497) that found that those convicted of CSAM very rarely (1.5%) reoffend with a child sexual abuse crime; they are more likely to reoffend with non-sexual crimes. This does not imply that such content should be tolerated, but rather that we should reflect on the approach currently used to combat this problem. Funding prioritizes strategies to combat individuals who have already committed the crime, which is the lowest level of crime prevention and ignores [other ways to combat this crime](https://irep.ntu.ac.uk/id/eprint/34860/1/12439_Winder.pdf): > «Consequently, it has been argued that effective prevention initiatives require the development and implementation of interventions at the primary, secondary, and tertiary levels. Adopted from the field of public health, these three levels of prevention have been described as follows: (1) Primary prevention involves wide-scale initiatives aimed at the general public and implemented before the occurrence of sexual violence to prevent even initial incidents of CSA (e.g. general crime deterrence, public education, adequate sex education in schools); (2) Secondary prevention involves more targeted interventions for those at-risk of engaging in CSA, which address issues known to increase the risk of offending (e.g. anonymous helplines for people with sexual interest in children); and (3) Tertiary prevention is a reactionary approach after a sexual offence has occurred, which aims to prevent sexual recidivism (e.g. treatment programs for those who have engaged in CSA) (Laws, 2000; Wortley & Smallbone, 2006).» (P. 3-4) > [@Factorial](#): > > Realistically, the child porn crusade is a greater threat to children than the people it prosecutes. Ironically, several people who have tried to prosecute child pornography are[the very victims they claim to help](https://www.usatoday.com/story/opinion/2016/01/10/perverting-teen-sexting-laws-criminal-felony-abbott-copening-column/77745364/). Instead of giving authorities more power to abuse their authority by funding measures that have not been proven effective in deterring crime ([for example](https://www.criminallegalnews.org/media/publications/The_Effectiveness_of_Sex_Offender_Registration_and_Notification_Sept._2021.pdf)), they should invest more in raising awareness among the general public and in [comprehensive sex education](https://www.sciencedirect.com/science/article/pii/S1054139X20304560) for young people so that they can navigate the web more safely: > This review found strong evidence for the effectiveness of child sex abuse prevention efforts in elementary school. Such programs typically use behavioral practice and role-play [80] and encourage parental involvement [81,82]. They teach about body ownership and children’s right to control their bodies [82] and about communication and self-protection [81,82]. A strong meta-analysis of 27 preschool through Grade 5 programs [80] and a systematic review of 24 K-5 programs [83] demonstrate significant effects on a wide range of outcomes, including behaviors in simulated at-risk situations. Another large systematic review concluded that, in general, parental involvement, opportunities for practice, repeated exposure, and sensitivity to developmental level were key characteristics of effective child sex abuse programs [81]. --- ## Post 33 by @win11.shading291 — 2025-09-14T00:30:14Z @flossposse @Factorial You should complain to the guardian. I agree that this is disappointing as most of the time, the Guardian seem to take facts seriously. > **[Editorial complaints and corrections | The Guardian](https://www.theguardian.com/info/complaints-and-corrections)** > > Latest Editorial complaints and corrections news, comment and analysis from the Guardian, the world's leading liberal voice --- ## Post 34 by @CTRLexe — 2025-09-14T00:39:42Z Do you know many pedos are on the clearnet? I was groomed on the clearnet. Plenty of pedos abuse kids on roblox and discord while those platforms do nothing about it. But no one cares because it doesn’t sound as “scary” as the dark web. These corporations steal your data, use the children as an “excuse”, then do nothing to protect those children when they get harmed. --- ## Post 35 by @dngray — 2025-09-14T05:44:46Z It’s sensationalist and designed to pull at emotional heart strings. You can usually tell with this topic when it’s vague, and makes assertions without an informed position. As others of said, the “darknet” is not most of the internet, and most of the pedophiles will take path of least resistance and be on mainstream things like regular social networks. I think this mission of trying close down and lock down everything so they will somehow be identified and excluded is never going to work. Others will of course argue that no cost should be spared, but is not a realistic position as all resources are limited. Nobody works for free in this world, and that includes law enforcement. I think at this point more focus should be on what to do with them once they’re identified and how it even happens in the first place. Many would say “lock them all up forever” but that costs money too and would only be shifting a problem out of sight, and out of mind. Those offenders are also someone’s children too and I’m not entirely convinced warehousing people and forgetting about them is a solution either. With young people, producing material of themselves or discovering something online a better approach would be harm reduction. Giving young people a criminal record for that serves absolutely no purpose but sending their lives potentially off-track and into isolation from the community, with the added potential for further offending. --- ## Post 36 by @Encounter5729 — 2025-09-16T22:35:06Z Sorry, but I lost faith in them since one of their journalist leaked a password of an archive with Snowden leak material. It was a genuine mistake but still.