IE like Crypto AG:
In 2020, it was revealed that the Swiss company, Crypto AG, which provided secure communications services to ~120 governments throughout the 20th century, was secretly ran by the CIA and West German Intelligence. The CIA and later NSA were able to read encrypted communications for many countries such as Saudi Arabia, Iran, Italy, Indonesia, Iraq, Libya, Jordan and South Korea.
i don’t think anyone here considers it a private service at all, but i’m almost certain cloudflare is a honeypot
Why are you so certain?
the biggest part is they’re doing way too much of the internet while being quite opaque. and their service is “too generous”, with free tiers, no ads. and the whole MITMing every traffic and serving from CDN architecture seems ideal for a honeypot to me.
even if cloudflare themselves don’t intend to be one, i’m pretty sure some three letter agency has backdoors to their systems.
Signal and Tor have both received huge amounts of US government funding, very suspicious.
DNS4EU and WiFi4EU.
Proxies and VPNs seem like the most obvious targets. They mostly prey on people who don’t understand the technical workings thereof (had my mom ask if she needed to get a VPN bc firefox opened on ad for theirs, claiming it enhanced privacy), and serve little benefit to people who are doing the kind of illegal activities that make governments take notice. They serve as a single point of compromise for anyone, and they work worldwide so that all your traffic can be monitored even when you’re on a different ISP/in a different country. It’s like the perfect MITM, and people are even willing to pay to have themselves monitored.
The truth is that at best they benefit people who only don’t want their network-provider watching, but don’t care who else may be. It’s the perfect setup for a 3-letter agency to just sit and monitor everything anyone does, waiting for someone who’s just a little too careless to access illegal content thinking they’re anonymous.
They are perfect for torrenting though. The kind of activity 3 letter agencies don’t want their spying to be disturbed for.
they benefit people who only don’t want their network-provider watching, but don’t care who else may be.
Just FYI: It’s not the network provide we have to worry about in my country. That is specific to the USA I believe.
Here they have “headhunters” that make a contract with a rights holder, torrent a file, write down the IP of someone who uploads a video to them, then legally request the name to the IP and send an invoice for about $2000. No three warnings or anything. And they are very good at sending legal officials to impound any of your valuable stuff in case you don’t pay.
Even other “illegal” activity like calling Israel an apartheid regime or supporting palestine or insulting your head of state might get you flagged by a three letter agency, but they won’t use official legal channels. There is a protection of the herd with VPN.
All of the “delete my information from data brokers” services IMO, especially the ones that advertise on YouTube. Always smelled fishy to me.
Either that or they’re just more data brokers trying to get exclusivity.
Reject Convenience did a pretty thorough rundown on what they’re doing: https://www.youtube.com/watch?v=iX3JT6q3AxA
It’s been a minute since I watched, but my key takeaways were that they just reach out to one type of broker which barely scratches the surface of the Data Economy iceberg, and since there’s no legal precedent outside of California and the EU, it’s purely up to the brokers to decide whether or not they want to comply.
So I think it’s probably more likely they really are just private companies preying on people’s anxieties about privacy and relative ignorance about the topic, rather than some kind of governmental conspiracy
Most likely all free vpns
Israeli actually, like express VPN
Maybe not a honeypot, but definitely too large for my taste by now: Proton. With Mail, VPN, password manager, file storage, AI and whatnot, it’s one ginormous basket to put all of your eggs into, hopping it’ll hold.
the owner is fine with fascism because fascism makes his product more lucrative
Did he say that? :o
not exactly. the more nuanced inspection of what he said was that donald trump’s plans to deregulate the tech industry he expected to benefit his company. however, that deregulation is in service of allowing more surveillance capitalism, environmental degredation, and worker mistreatment. the wording i provided is what that ultimately means as an analysis of how and why proton would make more money in that type of environment
Of course, nobody is going to have evidence here, if there was any the cover would be lifted. But one can guess chances here:
Proton: “Unlikely”… but there is a but. They never cater for the ultimate privacy and they make typical blunders of a company wanted to growth really fast. Now, that they want to be a behemoth in Privacy makes it more vulnerable to requests from law enforcement. Also, law enforcement and intelligence agencies have it easier to penetrate within Proton massive headcount growth.
Tuta: “Very Unlikely”. The people behind started very young and had a sustainable growth. The people are very visible (unlike Crypto AG) so least likely to be working for an “agency”.
Mullvad: “Very Unlikely”. I think their story is similar to Tuta (haven´t followed it that much though).
GrapheneOS: “Very Unlikely”. But in the last year I have raised some minor concerns, but I haven change my rating yet…
/e/: “Very Unlikely”. I know the dude behind for 2 decades, he wouldn´t. However, /e/ never claimed full privacy and from the beginning says he would comply 100% with “lawful” requests, but it is not a honeypot, not that would make much difference to an intelligence agency if they wanted it.
Signal: “Potentially”… yes, yes… audited, solid privacy code… but still does not make sense to me many aspects; financially solvent from day one, the extreme unquestioned massive and vast support from launching till today… if i have to bet in all of these providers, this platform would have been my take as potential compromised one. I still use it to communicate with family since I trust better than WhatsApp, but I would not use it for critical journalistic info.
Oh what are your minor concerns with GrapheneOS? I heard the head behind it is a little weird and paranoid, but honestly i think you kinda need to be for a project like that.
Signal requires to use phone number, which in many countries is legally required to be tied to your personal identity. Like the SMS provider must have a copy of your id card. You’re basically naked to the CIA when using Signal. Even if not like in the US they presumably mass collect SIM and location correlations for ID. For the life of me I do not understand how anyone can promote that shit.
So the “honeypot” of Signal is that the mainstream promotes it as IF it was a privacy focused app when it’s very glaringly obviously is not. So the effect is that it prevents market space and attention for other apps actually focused on privacy without requiring ID to sign up. It’s a bit like introducing sterile insects to prevent the spread of unwanted pests (= actually secure communication).
Mullvad: “Very Unlikely”. I think their story is similar to Tuta (haven´t followed it that much though).
https://tmctmt.com/posts/mullvad-exit-ips-as-a-fingerprinting-vector/
Great catch, is this still default behavior?
Thanks so much, this looks worth a deep look! Iĺl need time to interpret this properly.
Not a privacy app, but you should definitely not think anything said on discord is private in any sense whatsoever
I know your example is the opposite, but any service that is run and hosted in the US.
It’s one of the major issues with Signal.
Not to mention Graphite and Pegasus, Israeli spyware.
When parliaments have to inquire their own spy services, it’s a sign that these spy services must be disbanded, as they are becoming a deep state of their own, intimidating and harassing politicians. After all, if you can’t trust your own politicians, whom can you? And that’s problematic.
Disbanding those services and prohibiting any secret services from ever forming, would also regain a great deal of trust of society in each other. And that trust in turn, can foster society to advance for mankind.
You got that right.
Probably various VPNs on the market
Especially Israeli owned VPNs. Which seems to be most of them lately.
Oh yeah definitely
I always assume the more popular it is, the more likely it is of being compromised.
I have no idea if it’s the case, but I switched away from mullvad after seeing billboards and ads of it everywhere, even on city infrastructure like trains and buses.
If the company is owned by “Kape” its ikely a Israeli honeypot:
https://medium.com/illumination/vpns-the-privacy-trap-4aef67f39634
Kape’s portfolio includes ExpressVPN, acquired in 2021 for $936 million; CyberGhost, purchased in 2017; Private Internet Access, bought in 2019 for $127 million; and ZenMate.
Together, these services account for three of the six most popular VPN products globally, serving approximately 7.4 million paying subscribers.
Kape also owns VPNMentor and Wizcase, review platforms that rank VPN services — including Kape’s own products — for consumers seeking expert guidance.
Mullvad is very likely one of the few good ones. I’d suggest reevaluating it.
My trust in them was definitely shaken after the recent news about fingerprinting exit IPs: https://tmctmt.com/posts/mullvad-exit-ips-as-a-fingerprinting-vector/
They were very responsive but this seemed like a huge fuck-up to me, to the extent that I question whether it was purposeful.
Not sure who else to trust because other providers like Proton seem even worse
I have no reason to go back to it, and I switched away from it for the reasons mentioned: its grown very large, and has mainstream ads everywhere now.
if it makes you feel better i know an employee there and theyre a communist and say a lot of mullvad employees are lefties too, idk if they have a union or anything. nym vpn has chelsea manning backing it. not really a traditional vpn though its basically unfree tor that is not slow as balls, has the benefit of really good server coverage and few people blocking it. coolest thing is you can use a seedbox to route traffic to pay it down.
Especially the ones aggressively marketed, or noted as independent when they cannot give concrete evidence for whence their finances and ownership come. Always question and investigate, and make sure trusted people know you do so.
Most people only use vpn providers for streaming location hopping, torrenting, p*rn and on public networks. For day to day 24/7 use you are just trusting your VPN provider not to spy on your traffic instead of your ISP.
This thread basically illustrates the challenges for a beginner, such as myself.
I’ve been locked into the Google ecosystem for nearly two decades and am now trying to free myself.
I’d like to migrate to a hybrid solution that involves self-hosted NextCloud synchronized with a cloud provider that I can trust more than Google.
However:
Proton apparently makes false, or at least misleading, marketing claims and doesn’t fight a vast majority of its inbound government requests.
Tuta has been publicly accused by a member of the intelligence community of being a honeypot.
The rest of the email providers seem to implement even fewer protections, relative to these two.
So, what’s a guy to do?
Now, to be clear, I’m not saying that either of these companies are bad or that I believe that they’re actually honeypots. I’m just trying to illustrate the challenges faced by newcomers (and probably all of us).
While I’d prefer to absolutely maximize privacy and security on all fronts, given that my first goal is de-googling, I will probably start with Proton and NextCloud and re-evaluate from there, but I’m open to suggestions.
Thank you all – I really appreciate this community.
for email, the protocol itself is insecure by design. if using it for actual communication you should use something like pgp encryption on top. even proton receives your mails in plaintext, though they claim to store it encrypted afterwards.
get your own domain and use it instead of the provider’s domain, this way you can easily change email providers later on.
also btw, proton doesn’t support imap/pop (afaik)
Yes, I intend to use my own domain name when I switch.
For IMAP, it looks like there are bridges for both Proton and Tuta that I can run locally.
Email is a really tough one especially, because it wasn’t designed with security in mind, and of course even if you’re on a secure email service, 99% of the emails you send and receive are going to be with non-secure services hoovered up by google or AWS.
Anything is better than google at least.
Tbh for email I’d say don’t bother with privacy as it wasn’t meant to be private, as Dessalines said. If you care about data sovereignty (which is different to privacy, though often hand-in-hand), you can self-host email—it’s not as hard as it’s reputed to be. I’ve self-hosted my main email address for a couple years now and not had major hiccups. For the most part, after initial setup, it just runs. And if you’re daunted by configuring it, there are out-of-the-box solutions like Mailcow you can use. I’d only really recommend it if you already have a VPS/home lab/etc where you already self-host things.
I intend to do that but basically wanted to have an off site copy, for both backup and deliverability purposes.
I don’t have much in the way of privacy expectations for email, but I figure that Proton or Tuta are probably still safer than Google.
I self-host on a VPS, so my off-site copy is the VPS, and my on-site copy is the emails downloaded to my email clients.
I figure that Proton or Tuta are probably still safer than Google.
Define “safer”. If you are receiving unencrypted emails (which is the case in the vast majority of cases), there is nothing stopping Proton or Tuta from reading them. Fundamentally, if something arrives at a server unencrypted, the server can read it—nothing can be done about that.
If you’re exchanging e2ee emails, then it doesn’t matter if you use Google, because the body of the email can’t be read by Google. A lot of metadata is required to be unencrypted though (this is the case for Proton and Tuta too).
I don’t really see the benefit to using an email service like Proton or Tuta from a perspective of meaningful data privacy. If it were between e.g. Proton and Google I’d probably pick Proton to avoid my emails being used to serve me ads from Google, but I wouldn’t have any illusions about Proton being able to read unencrypted incoming mail.
Yes, I know and agree that the mail providers can read unencrypted email. I’d just rather use a provider that probably isn’t intentionally using it to build profiles about myself and others.
VPS/home lab
VPS is probably fine, hosting something this important on your own hardware sounds like a recipe for disaster though
the worse part is that; by the time security professionals’ tribal knowledge is known to the general public; it’s already outdated enough to keep you ensnared.
they say that you have to become your own lawyer to protect yourself and you have to become your own dentist/doctor to heal yourself; now you have to be your own secops to guard your information.
No company is in a position to resist lawful orders from government (not good orders, lawful).
It’s why every company that sells security makes a big show about planning to leave some western country when they say they’re gonna do mass surveillance. It’s all they can do.
Email is not secure and cannot be made secure.
Do not ever send anything through email that you rely on being private.
I’m certainly not suggesting that email providers should resist lawful orders, but if Proton complies with 89% of requests while Tuta complies with 25%, it suggests a difference in methodology, no?
It could, of course, be the case that the Swiss are just much more skilled at sending lawful requests relative to the Germans, but that seems unlikely.
So you have two different countries, two different sets of laws, and two different services with wildly different offerings.
You can’t really compare a drilled down percentage of compliance and reach the conclusion that there’s a difference in methodology under those conditions.
Just the much broader spectrum of services that proton offers makes it more likely that they will be in a position where they are required to comply with a larger portion of requests than tuta.
This is not intended to be a defense of proton, just a recognition that metrics are hard to take seriously in a comparison.
Tuta would make sense to me as a honeypot. Who called them out? Add it to the list of free providers I use that are just the CIA… In order to “anonymize” my social media profiles on their other sites lol
https://gizmodo.com/tuta-email-denies-connection-to-intelligence-services-1851022465
And again, I’m not saying that I believe this. I have no idea what to think. My original point was that it’s all very confusing to beginners.
Any VPN that isn’t actively being sued by world gov/agencies to try and get their data is suspicious.
Alternatively any VPN company with the ability to store data is untrustworthy.
Also every cryptocurrency that exsts.
How do you feel about Tailscale?
they were talking about proxy VPNs, whereas tailscale is for building actual virtual networks to connect your devices, which is a completely different thing (besides sharing the same approval foundation).
If you were to distrust tailscale (and you’re not simply self hosting headscale), an attacker might be able to access for otherwise non-public devices(’ ports), reroute/MitM your traffic and monitor which device connects to which.
Signal I think. I don’t mean that the end2end algorithm or messaging itself are itself unsafe, the algo has been shown to be secure. This is what people usually rebuke this with, with the reminder of Signal’s OSS nature.
The issue the servers and the social networking data that can be harvested. The server code only partially exists in public and we just have to trust that that is actually what is running on whatever AWS server without tampering and self hosting is nearly impossible in practice if technically possible and nobody does it. The social network data (who talks to who) is more valuable than the actual messages logs, which give a massive, but mainly useless datasets. Until LLMs, like 10-15 years ago they were basically impossible to parse for any useful info without using large quantities of eye pairs. Basically if you are an organizer, criminal, government, part of a hunted opposition, you will leak the whole core group structure of your org with attached phone numbers. Whoever with that data can then target their devices and persons with other means. Plus it’s literally built on top of CIA money. I think signal is totally safe and adequate for friends and family type of use, but not much else, but then all in all so is whatsapp, mostly since signal and Whattsapp share the same end to end algorithm.
Signal is def one, otherwise US government orgs like RFA and OTF wouldn’t be defending and pushing for it so hard in western privacy spaces, nor fund it.
Have a look at Deltachat
Its starting to make headway: FOSS, Decentralised and anyone who is tech inclined can setup their own Relay.
It’s funny how every poster who criticizes Signal inevitably makes a technical error. In your case, the claim that “Basically if you are an organizer, criminal, government, part of a hunted opposition, you will leak the whole core group structure of your org with attached phone numbers” entirely lacks basis. The Signal client - the OSS part we can and do control - does not divulge phone numbers.
You have this theory that Signal’s servers are storing communication records. (While there is no evidence to support this, it’s valuable to consider what they could do.) So the data that would be captured here is a network of hashed phone numbers and literally undecryptable messages. It’s impossible for the adversary to determine any phone numbers they don’t already know this way.
And since you can make a Signal account with a burner phone and create a “username”, even a known phone number becomes useless against targets who don’t want to be identified.
The US government could easily force google to put a compromised binary of signal on the google play store.
Good thing you don’t have to get the client app from the Google Store, then!
foreshadows…
All speculation. You gave them your phone number (which also means your real identity), so you should assume they have it. And because its a US-based company, it must adhere to US laws including key disclosure laws, which make it illegal for any signal employee to tell you that any US government agency has asked for this information.
https://en.wikipedia.org/wiki/National_security_letter
So the data that would be captured here is a network of hashed phone numbers and literally undecryptable messages
With this data you can build social networking graphs: who is talking to who, and when.
Also this is all the more suspect when you consider that US military / government agencies like OTF fund signal, and constantly try to push signal in privacy spaces.
which make it illegal for any signal employee to tell you that any US government agency has asked for this information.
That’s funny so this list of government data requests is just meaningless.
Yep it does. The Obama admin issued ~60 NSLs every single day, and I’m sure the number hasn’t decreased since then.
Signal doesn’t run in a vacuum. It’s main distribution platforms are app stores from Google and Apple. And most people are going to use stock smartphones from these two companies to sign up to Signal. But with them being under the same US jurisdiction, matching the two identities isn’t that far-fetched.
The parent companies of both OS platforms are well known to funnel data and notifications to the US government. It too had no evidence to support it, until they admitted it. There’s a setting for it now, but the person you’re talking to might not be doing the same, so it’s still out for profiling.
Other thing, they vehemently oppose F-Droid because “f-droid security flaws” bs, even though they can literally host their own repo for it without anyone else building their app. They would control every aspect of supply chain, but they didn’t.
Besides that, they make it very inconvenient to get it from elsewhere, even though they did the bare minimum to provide a standalone installer, after an outcry. And with those stripped down installers, you have to deal with inconsistent notifications, because no apple/google. And they never ever gave unified push a look. I wonder why? Are they a small indie company with just a couple of devs?
Signal protocol may be “secure”, but it’s only a part of a bigger picture.
It’s forced reliance on phone numbers, privacy averted platforms and unwillingness to work with opensource platforms and standards that lets it become decentralized and out of the hands of authoritarian government, leaves a lot to be desired.
Facebook’s whatsapp also uses the signal protocol, but would you call it private or secure after all that zuck has shown to do? Signal creator literally helped them implement it too. I wouldn’t touch a Facebook product with a 10 feet pole.
And now he’s helping them again encrypt Meta AI, whatever that means. Why is he working with one of the worst offenders of privacy?
If that doesn’t tell you these things are concerning, you do you.
The phone numbers being hashed doesn’t matter because of how small the input space is. A standard phone number is a country code plus 9 digits. If we assume that anybody looking at this information already knows what country the people they’re targeting is from, that means there is 1000 000 000 possible phone numbers to check for any hash. Even if the hash is extremely slow, and takes 1 second to compute on a strong CPU, that still only takes 1000 000 000 / (60 * 60 * 24) = 11574 days, or 31 years to compute on a single core. For any large organization (like, say, any government or any large tech company), getting 1000 cores to run the hashes in parallel would be quite simple, reducing the time it takes to have a complete hash list down to 11 days to get a complete database of all possible hashes. Hashing phone numbers is literally just a mild inconvenience.
Edit: Actually looking it up phone number formats vary quite a lot by country, but the point still stands.
The point is that they could. We are discussing honeypots here. They don’t advertise the fact if they are.
Be the phone numbers hashed/encrypted or not they will still get your ip. They are not routing anybody’s messages otherwise. Phone number is just more directly tied to a personal details, unless it’s a burner, but with burners you lose the account if you need to log in. Also you can set your phone number public, so it probably can be seen by the signal servers at some point. And what about discovery through phone number and like the actual sending of the signal confirmation code? How is any that suppose to work if the servers don’t know your actual phone number? And your anonymity trick only works if everybody you talk to does it, which they don’t. If they want to profile you they can profile you directly or through the people you talk with. If the people you are trying to hide from don’t care about getting message logs and just association with some group is punishable or can lead to punishment or death then tough luck.
And you miss the main point. practically speaking you cant self host a signal server, therefore you can’t trust it fully (in a way ‘fully’ matters anyway). if you do it’s unsupported and not recommended and you probably need a custom client to access it. That added with it being under American jurisdictions, and Signal starting as a spook project should really set off alarm bells.
