Wasabi wallet data privacy questions

[o_e_l_e_o]

I've removed it.

Thanks!

I'd like to understand more about coinjoins for widening my learning. I own tiny amount of bitcoin so won't be coinjoining frequently or in large volumes. I want to learn how it works.

There are a few topics on this forum which are a good starting place.

Here is the post by Greg Maxwell introducing the concept of coinjoin for the first time: CoinJoin: Bitcoin privacy for the real world
And here is the post by Chris Belcher launching JoinMarket: [ANN] Joinmarket - Coinjoin that people will actually use

It's a heavy read because of details given in long questions & answers but I'll try going over it.

The bottom line is of the three current main coinjoin implementations - JoinMarket, Whirlpool, Wasabi - Wasabi is the only one cooperating with blockchain analysis, the only one supporting mass surveillance, and the only one implementing blacklists and censorship, not to mention suffering from address reuse and other flaws. There is literally no reason to choose Wasabi over either Whirlpool or JoinMarket.

Thanks o_e_l_e_o for amazing advice.

No problem!

[Kruw]

@kruw thanks I'll ask other questions after analysing your replies. It doesn't look like zkSNACKs are keeping info which links users after mixes but it's about trusting what they're saying.

There's no trust in zkSNACKs necessary - Wasabi is completely open source so you can verify for yourself that no identifiable data is ever sent to any third parties, including coinjoin coordinators such as zkSNACKs.

Wasabi wallet being marketed as a privacy solution while simultaneously funding blockchain analysis is contradicting. I haven't seen data they're doing excessive bad things. Analysing incoming coins for naughty status isn't the biggest crime if it's designed to stop crime. If my wallet's hacked coins were sent to any address on its way to Wasabi for mixing I'd be happy if they succeeded in stopping coinjoins so I'm not getting why that makes people upset.

Yeah, it's quite strange to see o_e_l_e_o be so passionately furious at the idea of a business refusing to make stolen money private that he would misdirect that anger at an open source project that allows anyone to run a  competing business.  If he hates zkSNACKs so much, why isn't he simply running his own WabiSabi coordinator by copying and pasting their open source code and making them go out of business by offering their service for free instead of charging any fees?...

Which data's being stored by zkSNACKs in this transaction if I'm using Wasabi wallet.

o_e_l_e_o sends 0.1 btc to arabspaceship123
arabspaceship123 sends 0.1 btc to Wasabi wallet
Wasabi wallet sends mixed 0.1 btc to arabspaceship123
arabspaceship123 sends 0.05 btc to o_e_l_e_o
arabspaceship123 sends 0.05 btc to theymos

The data stored by zkSNACKs is the exact same data that is stored on everyone else's full node when these transactions are confirmed in blocks.  zkSNACKs has no knowledge about who owns coins that have been mixed.

In your opinion what's right to do if your wallet with 1 btc gets hacked. On blockchain you trace funds were sent to wallet bc1xx so you report it to police. A month later the coins moved to Binance. Soon your coins will be returned because they've been seized after blockchain analysis. That's happening now with exchanges so it can't be a bad process reuniting owners with stolen coins.

Scanning incoming UTXOs in Wasabi wallet it's the same as Binance or other exchanges doing it. What they're doing with data isn't clear. If it's being misused for unfair data collecting after it's completed its purpose it shouldn't be used. I don't trust any companies to voluntarily delete data after if they're able to profit from it.

It's not quite the same as an exchange because there's no way that stolen coins can be seized by a coordinator, coinjoins are non custodial.  If a coordinator realizes coins are stolen, they can only refuse to include them in coinjoin transactions they coordinate.

Wasabi pay Coinfirm to investigate the output you are registering for coinjoin. So Coinfirm will absolutely be looking at the history of that output and seeing where it came from. If any of the previous addresses have ever been linked to an identity (such as via KYC, via addresses being shared publicly, via connecting to third party servers, via other transaction heuristics, etc.) then that will be identified and Coinfirm will be storing, sharing, and selling, that information on to other third parties.

This is completely wrong.  Wasabi coinjoin coordinators do not have any knowledge of who coinjoin outputs belong to.  You would be correct backdoors such as those in Samourai's coinjoin coordinator allow tracking the outputs that were created from coinjoin inputs due to xpub addresses and IP addresses being leaked by the Samourai or Sparrow client, but Wasabi clients automatically assume any coordinator is malicious and never reveal ownership of coinjoin outputs to them thanks to using compact block filters and Tor by default.

They're storing the address used by Satoshi in tx1 before coinjoins but aren't storing tx details after mixing so they can't link anything between tx2-7.

Yes, this description is correct.

[n0nce]

Which data gets transmitted to zkSNACKs when Wasabi wallet opens in Win 11?

Nobody knows this information, because Wasabi won't tell.

That's not really true. As long as you download the source files from their GitHub page and compile it yourself, you can check the code that runs on your own device.
I.e., you can read for yourself what information is sent out and what stays local.

However, the tricky part is that there's no way to tell what they are doing with the information the application does send to their servers.
Some information has to be sent for it to function, and it can definitely be used maliciously. Many users here agree with me, that after the statements Wasabi officially made, we know for a fact that they do act maliciously in terms of Bitcoin's original vision of freeing people from censorship and surveillance.

[o_e_l_e_o]

Yeah, it's quite strange to see o_e_l_e_o be so passionately furious at the idea of a business refusing to make stolen money private that he would misdirect that anger at an open source project that allows anyone to run a  competing business.

The "nothing to fear, nothing to hide" argument is so monumentally stupid and so widely discredited that to use it in this context you are either an idiot or actively malicious.

If he hates zkSNACKs so much, why isn't he simply running his own WabiSabi coordinator by copying and pasting their open source code and making them go out of business by offering their service for free instead of charging any fees?...

Because, as I've given you evidence of at least a dozen times and you have completely ignored at least a dozen times, Wasabi coinjoins are deeply flawed and result in outputs being linked to inputs due to various factors such as UTXO sizes and address reuse. Why would I want to run an inferior coinjoin coordinator when I can just use a much superior product such as JoinMarket or Whirlpool?

This is completely wrong.  Wasabi coinjoin coordinators do not have any knowledge of who coinjoin outputs belong to.

Cool strawman. That's not what I said at all. I said Coinfirm will be analyzing the inputs to ascertain their history and see to whom they belong, not Wasabi. You were the one here which conflated Wasabi and Coinfirm as being the same entity, which is frankly pretty hilarious.

we know for a fact that they do act maliciously in terms of Bitcoin's original vision of freeing people from censorship and surveillance.

This. Funding blockchain analysis is malicious. Gaslighting people in to believing that funding blockchain analysis is somehow in their best interests is just downright evil.

[Cricktor]

My wallet tests were on Win 11 using good security solutions. It isn't the safest but it's the biggest OS.

The biggest OS (I assume you mean largest marketshare or user base) is also the preferred playground for malware creators. I don't want to judge and/or compare the security of Windows vs. Linux vs. MacOS. But because Windows as OS is the preferred target for malware creators I wouldn't choose it as platform for crypto coin activities.

And as data privacy questions are a concern of the OP and as has been said here already that in this respect Windows is really terrible, I wonder why the OP sticks to Windows. You have at least two big arguments to stay away from Windows. (I acknowledge that OP is looking for some better solution with that.)

I prefer to use coinjoins with Sparrow wallet or JoinMarket (I don't care if coinjoins take a longer time; I can separate short-term from long-term coin use; what I want to make more private by coinjoining are rarely coins to be in a hurry with).

The controversy around Wasabi is pretty much heated and somewhat stuck in my opinion. I lack the time and coding knowledge to do my own research and code audit on Wasabi. And as there are alternatives, my choice is to stay away from Wasabi. YMMV...

[o_e_l_e_o]

I prefer to use coinjoins with Sparrow wallet or JoinMarket (I don't care if coinjoins take a longer time; I can separate short-term from long-term coin use; what I want to make more private by coinjoining are rarely coins to be in a hurry with).

I hit this milestone in Sparrow a few weeks ago, and decided to screenshot it for posterity:



I've added a few more to that count since then, and I've got a couple of such inputs in the same ballpark. It's turned in to an experiment now where I will just refuse to spend at least one of these outputs unless absolutely necessary just to see how high we can go. With every remix after the first completely free, of course.

I lack the time and coding knowledge to do my own research and code audit on Wasabi.

You don't need to. They openly admit they are cooperating with blockchain analysis and directly funding state sanctioned surveillance and censorship. That alone is more than enough to mean nobody in the right mind should ever use Wasabi.

[arabspaceship123]

They're storing the address used by Satoshi in tx1 before coinjoins but aren't storing tx details after mixing so they can't link anything between tx2-7.

Yes, this description is correct.

Kruw thanks you've replied when asked. You've settled questions bugging me but you're defending Wasabi Wallet when zkSNACKs isn't trusted. They're using blockchain analysis for coinjoins so tech forum users won't use it. If trust's lost it isn't easy regaining it.

However, the tricky part is that there's no way to tell what they are doing with the information the application does send to their servers.
Some information has to be sent for it to function, and it can definitely be used maliciously. Many users here agree with me, that after the statements Wasabi officially made, we know for a fact that they do act maliciously in terms of Bitcoin's original vision of freeing people from censorship and surveillance.

If the information's anonymous or untraceable when it's sent to their servers it shouldn't matter what they're doing with it. If it's linked to users data it's inappropriate. Did tech companies run source code tests to examine information being sent to zkSNACKs ?

Kruw answered questions with strong belief in what he's said so I'll give credit on his effort. He's defending Wasabi wallet when he knows he won't convince determined users about protecting Satoshi's words but he doesn't avoid answering.

The biggest OS (I assume you mean largest marketshare or user base) is also the preferred playground for malware creators. I don't want to judge and/or compare the security of Windows vs. Linux vs. MacOS. But because Windows as OS is the preferred target for malware creators I wouldn't choose it as platform for crypto coin activities.

The biggest OS by market share. That's why malware creators target Windows because of it's popularity.

[Kruw]

Kruw thanks you've replied when asked. You've settled questions bugging me but you're defending Wasabi Wallet when zkSNACKs isn't trusted. They're using blockchain analysis for coinjoins so tech forum users won't use it. If trust's lost it isn't easy regaining it.

That's the entire point of Wasabi:  You don't have to trust any coinjoin coordinator you choose since no information is ever provided to them.  Tor protects your IP address and block filters protect your xpub address.

[o_e_l_e_o]

You don't have to trust any coinjoin coordinator you choose since no information is ever provided to them.

Except you do have to trust zkSNACKs not to censor you after they are done paying blockchain analysis companies to spy on you.

There are other blockchain coinjoin implementations which do not have this weakness. Any sensible person will use one of them instead.

Edit: typo.

[arabspaceship123]

Kruw I've used a previous example so here's a newer. Rejected coins return to original address that's perfect if it's Bitcoin Core or Electrum wallets types. What happens next shouldn't include zkSNACKs or Coinfirm keeping tabs on my prior history UTXOs in Wasabi wallet or any future ones. As the receiver user I'd like to know will zkSNACKs or Coinfirm put eyes for following me or connect me to the naughty coins? It's important for understanding what Coinfirm does with data if they've analysed naughty coins.

tx1 Satoshi uses Binance sends 0.1 btc to arabspaceship123 Electrum wallet
tx2 arabspaceship123 Electrum wallet sends 0.1 btc to Wasabi wallet for mix
tx3 Wasabi wallet use Coinfirm for analysis. My 0.1 btc considered naughty so banned
tx4 Wasabi wallet returns 0.1 btc to arabspaceship123 Electrum wallet

Naughty coins will be sent back to the same address but the receiver won't get them if they've originated from mixers. What happens to banned naughty coins if they're sent from single use addresses in Sinbad Coinomize or Whirlwind?

tx1 Satoshi uses Binance sends 0.1 btc to arabspaceship123 Electrum wallet
tx2 arabspaceship123 Electrum wallet sends 0.1 btc to Sinbad wallet for mix
tx3 after mix Sinbad sends 0.1 btc to Wasabi wallet for mix
tx4 Coinfirm for analysis. My 0.1 btc considered naughty so banned
tx5 Wasabi wallet use Coinfirm for analysis. My 0.1 btc considered naughty so banned
tx6 Wasabi return 0.1 btc to Sinbad address but I won't get received funds

Kruw thanks you've replied when asked. You've settled questions bugging me but you're defending Wasabi Wallet when zkSNACKs isn't trusted. They're using blockchain analysis for coinjoins so tech forum users won't use it. If trust's lost it isn't easy regaining it.

That's the entire point of Wasabi:  You don't have to trust any coinjoin coordinator you choose since no information is ever provided to them.  Tor protects your IP address and block filters protect your xpub address.

[o_e_l_e_o]

-snip-

Wasabi won't automatically send your coins anywhere. If you deposit them to your Wasabi wallet and zkSNACKs decides you are a naughty little kid who is not allowed to coinjoin, then your coins just sit in your Wasabi wallet until you send them somewhere else. They won't automatically be returned to the previous address.

As the receiver user I'd like to know will zkSNACKs or Coinfirm put eyes for following me or connect me to the naughty coins?

He can't answer this because he doesn't know. Nobody except Coinfirm know what their secret analysis involves or know just how far backwards or forwards they are keeping track of your coins. Suffice to say, however, it's more than a single transaction, and they will be using all the data they have at their disposable to deanonymize and track your coins. This is what you subject yourself to if you use Wasabi.

[arabspaceship123]

-snip-

Wasabi won't automatically send your coins anywhere. If you deposit them to your Wasabi wallet and zkSNACKs decides you are a naughty little kid who is not allowed to coinjoin, then your coins just sit in your Wasabi wallet until you send them somewhere else. They won't automatically be returned to the previous address.

Before I didn't understand what happened to rejected naughty coins so I assumed the negative. It's better if there isn't a risk of coins beings sent back to addresses. If funds sit in Wasabi wallet it's owners can send them when it's convenient.

As the receiver user I'd like to know will zkSNACKs or Coinfirm put eyes for following me or connect me to the naughty coins?

He can't answer this because he doesn't know. Nobody except Coinfirm know what their secret analysis involves or know just how far backwards or forwards they are keeping track of your coins. Suffice to say, however, it's more than a single transaction, and they will be using all the data they have at their disposable to deanonymize and track your coins. This is what you subject yourself to if you use Wasabi.

How's Wasabi wallet connecting with Coinfirm for data analysis and what's Coinfirm doing with the info after they've decided coinjoin status? Kruw said Coinfirm secret analysis isn't connecting coinjoin outputs so does it mean it isn't bad if it's true?

[o_e_l_e_o]

How's Wasabi wallet connecting with Coinfirm for data analysis

Whatever coins you try to coinjoin, Wasabi pays Coinfirm to spy on those inputs and decide whether or not they are "naughty".

and what's Coinfirm doing with the info after they've decided coinjoin status?

We don't know, and Wasabi won't tell us. I suspect Kruw doesn't know this either - the knowledge is probably limited to the top 2 or 3 devs at Wasabi. But given that all blockchain analysis companies only exist to gather data and then sell and share that data with third parties, I would be very surprised if Coinfirm don't use this data that Wasabi pay them to gather in a variety of other nefarious ways.

[Kruw]

You don't have to trust any coinjoin coordinator you choose since no information is ever provided to them.

Except you do have to trust zkSNACKs not to censor you after they are done paying blockchain analysis companies to spy on you.

There are other blockchain implementations which do not have this weakness. Any sensible person will use one of them instead.

What "weakness"?  What do you mean "blockchain implementation"?  You are making no sense.

Rejected coins return to original address that's perfect if it's Bitcoin Core or Electrum wallets types.

Not quite:  Rejected coins never leave their original addresses at all.  No fees is paid, no coins are spent, no data is revealed.  Being rejected from a coinjoin is completely harmless to the user, they can simply register with a different coordinator with absolutely no consequences whatsoever.

What happens next shouldn't include zkSNACKs or Coinfirm keeping tabs on my prior history UTXOs in Wasabi wallet or any future ones. As the receiver user I'd like to know will zkSNACKs or Coinfirm put eyes for following me or connect me to the naughty coins? It's important for understanding what Coinfirm does with data if they've analysed naughty coins.

zkSNACKs or Coinfirm cannot follow you or connect you to the naughty coins because your IP address is protected by Tor.  They cannot connect the UTXO to any other coins in your wallet because your xpub address is protected by your client's use of block filters (https://bips.xyz/158).

This is the most important difference between Wasabi and other clients like Samourai/Sparrow:  Wasabi's developers and coordinators NEVER gain any data from you, whereas the Whirlpool coinjoin coordinator collects your data and will notify you when it gets turned over to the government:  https://samouraiwallet.com/canary

and what's Coinfirm doing with the info after they've decided coinjoin status?

We don't know, and Wasabi won't tell us. I suspect Kruw doesn't know this either - the knowledge is probably limited to the top 2 or 3 devs at Wasabi. But given that all blockchain analysis companies only exist to gather data and then sell and share that data with third parties, I would be very surprised if Coinfirm don't use this data that Wasabi pay them to gather in a variety of other nefarious ways.

You do know because Wasabi is open source, just like Bitcoin itself.  You can verify for yourself that no identifiable data is ever leaked by Wasabi clients to its developers or any coordinators.  Wasabi is very proud about its zero data collection policy, yet you continually say that "you don't know" what data is collected.

The answer is ZERO.

Do zkSNACKs store any data to link users from received<>sent addresses after mixing?

We don't know and they won't tell us.

Which info does zkSNACKs share with blockchain companies is it limited to scanning incoming mixes for analysis?

We don't know and they won't tell us.

You do know, the answer is "none" because Wasabi does not reveal any data to them.  zkSNACKs is very proud of their zero data collection policy and talks about it all the time, how could you miss it?

https://twitter.com/wasabiwallet/status/1678694963712802816
https://blog.wasabiwallet.io/zksnacks-means-zero-knowledge/

[arabspaceship123]

How's Wasabi wallet connecting with Coinfirm for data analysis

Whatever coins you try to coinjoin, Wasabi pays Coinfirm to spy on those inputs and decide whether or not they are "naughty".

How are zkSNACKs servers connected to Coinfirm to check incoming UTXOs. They haven't told how it's processed. If mixers wanted to reject stolen coins wouldn't it be convenient for hacked wallets addresses be posted in a central site. It stops blockchain analysers in direct communication with mixers.

and what's Coinfirm doing with the info after they've decided coinjoin status?

We don't know, and Wasabi won't tell us. I suspect Kruw doesn't know this either - the knowledge is probably limited to the top 2 or 3 devs at Wasabi. But given that all blockchain analysis companies only exist to gather data and then sell and share that data with third parties, I would be very surprised if Coinfirm don't use this data that Wasabi pay them to gather in a variety of other nefarious ways.

When a company gets data it means they can make money from selling it. If it's sitting with data collection companies you're sure it's going to be sold. They way Kruw's explained Coinfirm isn't doing anything with the data except checking for stolen coins. Should zkSNACKs develop a different strategy to check for stolen coins?

[o_e_l_e_o]

What "weakness"?

The weakness that if you use Wasabi, you can no longer use bitcoin as it was designed - peer to peer with no third parties. You can't do this because you have to ask Wasabi to ask Coinfirm for permission to spend your coins in the way you want. If Coinfirm decide you are a naughty child, then you are censored. This is obviously a significant weakness compared to other coinjoin implementations.

What do you mean "blockchain implementation"?

A typo. Fixed.

You do know because Wasabi is open source, just like Bitcoin itself.

He was specifically asking what Coinfirm do with the data Wasabi pay them to gather on your UTXOs. Whether or not Wasabi is open source is utterly irrelevant to this question. As usual you are answering completely different questions to what is actually being asked because you aren't honest enough to give the uncomfortable answers.

Coinfirm are a blockchain analysis company whose entire purposes and entire existence depends on gathering and selling data. Any data Wasabi pay them to gather will be used for any other analysis they want and sold to any other third parties they want.

[Medusah]

Being rejected from a coinjoin is completely harmless to the user, they can simply register with a different coordinator with absolutely no consequences whatsoever.

Being rejected from a coinjoin means being rejected from having your privacy protected, which is rather the opposite of harmless.  You are harming the people who might have not done anything and whose coins are arbitrarily subjected to crime by denying to anonymize their coins.  Sure, they can go and try out in another software, as JoinMarket, but you do harm the reputation of Bitcoin privacy software.

As for the "simply register with a different coordinator", how is that going?  Is there even one that is running separately from zkSNACKs?

[arabspaceship123]

When zkSNACKs made the deal they should've guaranteed info wouldn't be used for other analysis or sold. What's in the fine print between zkSNACKs & Coinfirm matters to users. Is it available? Kruw's saying Wasabi wallet isn't taking data collection but he hasn't said what Coinfirm's doing with data.

You do know because Wasabi is open source, just like Bitcoin itself.

He was specifically asking what Coinfirm do with the data Wasabi pay them to gather on your UTXOs. Whether or not Wasabi is open source is utterly irrelevant to this question. As usual you are answering completely different questions to what is actually being asked because you aren't honest enough to give the uncomfortable answers.

Coinfirm are a blockchain analysis company whose entire purposes and entire existence depends on gathering and selling data. Any data Wasabi pay them to gather will be used for any other analysis they want and sold to any other third parties they want.

[o_e_l_e_o]

What's in the fine print between zkSNACKs & Coinfirm matters to users. Is it available?

No, it's secret. Wasabi won't even tell you something as basic as why your coins are being censored. They certainly aren't going to tell you what is in the small print of their agreements.

Kruw's saying Wasabi wallet isn't taking data collection but he hasn't said what Coinfirm's doing with data.

He is deliberately not answering that because either he doesn't know, or he does know and it's awful as I described above. If he does respond to this direct question, I fully expect him just to repeat his same nonsense sound bites.

[Medusah]

What's in the fine print between zkSNACKs & Coinfirm matters to users. Is it available?

We don't even know for sure it's Coinfirm.  The community in this place and in Twitter had reached in that conclusion, but they have not made it official.

Kruw's saying Wasabi wallet isn't taking data collection but he hasn't said what Coinfirm's doing with data.

At this point, I think Kruw is only defending the Wasabi project from bias, because he is a contributor.