Your thoughts on bitcoinfog.com in KYC/AML concerns?

[Tori_Stanwood]

People here have got me thinking seriously about using bitcoinfog for part of my chain of transactions. Since things are getting ready to happen this week I want to ask about your personal use of bitcoinfog.com.

Going through Google the last couple of days I have found that the people who love the fog love it and those that hate it just outright hate it. The biggest complaint I have found online is that the fog takes so long to do it cleaning of the coins. Many will post on forums when they think it takes too long to clean the coins a bit and start complaining about getting ripped. One guy I spoke to said it took him over a week to have his coin post. But my thinking is that the service they provide as far as splitting the coin up and all is a great service thus you should be very willing to give it the time to do it's job if KYC/AML is your main concern.

I would like to hear your thoughts however on bitcoinfog.com. Has anyone had a bad experience with using it? What should be my concerns if I use it? How best should I use it? One person told me to make several accounts and post a little to each account. I have read where a lot of people have complained about accounts getting deleted and thus the funds are lost. Is this something someone knows first hand about?

Thank you all for any info you can provide. Also to make sure you understand my concerns here. It looks like I will have around $18.000 to move through the system. This will help put into perspective my concerns here to try and get this all right and not make any mistakes I can't recover from. I fully expect to pay out 10% to clear this up before I turn it into cash. Thanks.

[Quickseller]

A long time before I got really involved in bitcoin, and involved in the forum, I would lurk in the forums of Silk Road (I never signed up for a SR account, or even created a forum account). It appears that bitcoin fog is a favorite for users of the dark markets to use to mix their coins. Although there are a lot of non-illegal ways to use them to mix your coins. IMO a likely reason why people who use dark markets like bitcoinfog is because it uses tor which probably gives them the illusion of privacy.

It has been a while since I have even looked into bitcoin fog, so I do not have the most up-to-date information as to any problems they might be having, how they work, or even if it is a good idea to use them now. I know that, at least how they would work in the past is that they would have a number of wallets (*for the purposes of this post, by 'wallets', I mean a group of addresses that may be associated with each other by spending inputs from multiple addresses - you would likely not be able to tell what all the addresses are without access to their database and/or private key files) that would be kept separate from each other. When you would fund your account, your bitcoin would fund one of their wallets, and when you go to withdraw your funds the funds would come from another of their wallets. When one of their wallets would get "low" they would transfer funds from one wallet to another. They would also have a time delay from the time that you sent the funds to the time your account would get credited, in order to prevent traffic analysis from you accessing tor and/or their .onion site and of the time that particular transactions funded their potentially known addresses. They would also force you to wait a somewhat random amount of time before you can withdraw your funds to prevent blockchain analysis of similar sized transactions.....they would also charge a somewhat random fee for a similar reason.

It should be noted however that the US government was able to recently able to take down a large number of tor sites, and although it was not publicly listed as a site that was taken does, that does not mean that it has not been compromised by the US government (which would at the very least result in the point being defeated to using their service for a good part - you would lose privacy).

Your post somewhat implies that you are planning on sending you coins through bitcoinfog multiple times, and this would probably be a bad idea. If you want to do something like this then you would probably want to use multiple mixers to allow you to have additional privacy. The only other mixer out there that is not a scam is bitmixer.io (although it is possible that they are also no longer legit), and it is possible they are also somewhat compromised.

If you are looking for privacy then I would suggest that you sell your bitcoin for DRK (dark coin) on an exchange, transfer your DRK to a DRK address that you control (repeat this a few times) and then finally sell your DRK for bitcoin on another exchange.

If you need any help with this then let me know

[shogdite]

I've used Bitcoinfog quite a lot over the years, never had that many issues apart from the occasional delayed deposit/withdrawal. Only had to contact their support once and they got back to me within a few hours so on the whole I'd recommend them. Personally if I were mixing that amount of btc I would do it in smaller amounts then exchange through several alts at various exchanges.


Might want to check their thread for more info:  https://bitcointalk.org/index.php?topic=50037.0

[Tori_Stanwood]

Thanks, and hope to get more feedback.

I am not depending on any one service or services reputation as law. I cannot afford to do that. But knowing the most I can while going in and then developing my own method of mixing and moving is the only way I think I can avoid or have the best option to avoid pitfall that know one knows about yet. I am always expecting the most unexpected and ludicrous of out comes that I have been told I was totally going overboard. Everything I do at each step requires my thinking that I do not know the situation on the other end no matter what I think I know. I can't trust anything outside of my control totally so I have to do what I do in a manner to protect myself from the most unlikely of events. I like my life and my freedom.

My thinking is to be as unstoppable and as untraceable as I can manage I need to do something like the below.

Using ip that is hundreds of miles away. ---> Get funds in coin ----> from my encrypted wallet to mixing service (A) lump sum so I get control the quickest. Withdraw from mixing service (A) from another IP in several small amounts over a long period of time to encrypted wallet (b - this could be several different wallets in different amounts) ----> (c) now transfer some to another mixing service while transferring some into another altcoin. Over time retransfer the rest into another altcoin. (d) move some from the other altcoin back into bitcoin mixing at different amounts and intervals. (e) move varying amounts at varying times into another mixer. (f) now move small amounts from mixer at varying intervals and amounts into another encrypted wallet. (g) move varying amounts at varying times into varying local bit coin accounts in varying locales to move for cash. Then contact this site and make a donation. Keep in mind I plan on doing each step using IP's separated by many miles and using the Tails Operating System.

I am still researching and will be up till the very last point of doing this so if someone wants to cut holes in my plan then go right ahead. Make me consider something I have yet to think about whether negative or positive. Break it down, cut it up, praise it, whatever you can give as far as input is only a good thing.

[Giggs]

Why don't you try bitmixer instead? I've not used bitcoinfog so I can't really comment on their service but I've used bitmixer a few times and never had any problems with their service.

[DeathAndTaxes]

Why would anyone hand over ownership/control of large sums of funds to an anonymous entity?  When centralized anonymous entities end up with massive amounts of coins then tend to disappear and 'everyone' seems surprised it happened.  The same level of privacy enhancing can be performed using coinjoin without needing to trust that a centralized service will not abscond with your funds.

[Tori_Stanwood]

I am not opposed to coinjoin and have been looking into it as well.  https://bitcointalk.org/index.php?topic=279249.0

Still working on my understanding of it also but looks good. I have been trying to figure out the benefit in coinjoin over what I am thinking of doing other than my current thinking is that it will cost me more to do what I have been thinking. Not sure if I would be safer with one or the other option.

[Quickseller]

Why would anyone hand over ownership/control of large sums of funds to an anonymous entity?  When centralized anonymous entities end up with massive amounts of coins then tend to disappear and 'everyone' seems surprised it happened.  The same level of privacy enhancing can be performed using coinjoin without needing to trust that a centralized service will not abscond with your funds.

blockchain.info's shared coin uses this feature. There have been some researchers/academics that have been able to trace the "path" of inputs throughout this service. The reason being that roughly the same size output will eventually need to end up at an address that the original input was. All of the transactions that take your coins from address A to address B are generally confirmed in the same block, so an adversary would almost always only need to look at one block in order to determine the address where your funds ended up. I would say that this would likely involve privacy against an adversary with only very minimal resources/interest and would only delay detection from an adversary with more resources

[DeathAndTaxes]

Blockchain.info shared send does not use coinjoin.  It is another inferior centralized service.   Coinjoin uses standardized outputs. So if bob puts in a 5 BTC input and joe puts in a 2 BTC input and there are seven 1 BTC outputs which ones belong to Bob and Joe. 

[Quickseller]

Blockchain.info shared send does not use coinjoin.  It is another inferior centralized service.   Coinjoin uses standardized outputs. So if bob puts in a 5 BTC input and joe puts in a 2 BTC input and there are seven 1 BTC outputs which ones belong to Bob and Joe. 

Well if bob has 5 BTC that he wants to mask the ultimate source of the inputs, and he receives five 1 BTC outputs to five different (previously unused) outputs then, based on the fact that I know that a very high percentage of people that use bitcoin are not able to properly manage their security/privacy/ect., I would say that there is a good chance that he would eventually sign/broadcast a tx that contains inputs from the 5 output addresses that he received funds from. Even if Jane did not make the same mistake then an adversary could determine which of the outputs belong to Jane based on Bob's mistake. If you were to have a larger set of people participating in a transaction (say 5 or 6 people, with a total of 40 or so BTC being transacted, with 1 BTC outputs) then it would still be possible to determine which outputs belong to who if enough participants make a similar or other mistakes that would allow an adversary to determine which outputs belong to the person they are looking for (with a reasonable amount of certainty).

Any time you want to participate in coinjoin or any similar variation then you will need to use somewhat of a centralized service/meeting place that uses some level of automation. Otherwise it would take too long to find a sufficient number of participants and you would waste a lot of time in waiting for everyone to sign their part of the tx, and would likely have many instances when one or more of the participants will not sign the tx. As an experiment, prior to writing this post, I used blockchain.info's shared send (which should give a rough approximation as to how long a coinjoin tx should take), and it took roughly 2 minutes and 30 seconds from the time that I 'authorized' the transaction on blockchain.info and the time that my instance of armory picked up the output transaction, while this probably would not seem like very long, I would estimate that if you were to broadcast a transaction to an address that my instance of armory is "watching" then it would take my instance of armory only seconds to pick up the transaction.

Although it is somewhat no longer relevant, I did take the time to find the article I was referencing in my previous post regrading the privacy of blockchain.info's shared send being broken, but for some reason I did not post it. So here it is

[DeathAndTaxes]

I would say that there is a good chance that he would eventually sign/broadcast a tx that contains inputs from the 5 output addresses that he received funds from.

Why would you conclude that?  Based on that logic all 7 BTC from the first coinjoin belong to bob.  There were two inputs to the same txn one of which (the 5 BTC txn) you know belongs to Bob.  Thus both inputs and all the outputs belong to Bob. Right? Also this is a highly simplistic example. In reality a coinjoin transaction will involve many participants.  Gmaxwell also points out in the coinjoin introduction that properly done it should involve multiple rounds.  So Bob should take some of those coinjoin outputs and use them for another coinjoin transaction and those outputs and use them for another coinjoin transaction.

Also the naive solution is to use a centralized server but that isn't a requirement.  It is like saying a system like Bitcoin need a centralized server to hold the ledger or that peer to peer file sharing still requires a centralized server for tracking.  Making a centralized system is easier so they often come first but in time coinjoin will grow to have a decentralized peer finding capability as well.  The peer setting up the transaction would still know the input/output mapping but using a two step commit and blinded output signing it would be possible to make even the organizing peer blind to the final transaction map.  Combine that with multiple rounds and you have a very robust solution.


You point out the 'flaws' of decentralized trustless systems and forget the obvious.  If I worked for a three letter agency and I mixers would make my job difficult in the future, I would be adovating now to either create or infiltrate centralized 'secure and private' mixer services to be honeypots for compromising privacy.  It would make things very opaque from the outside but at the same time make things very transparent for those who have access to the logs.  Provide me cryptographic proof that a mixer doesn't keep logs.  Even a mixer started with the best intentions is subject to infiltration by coercion, bribery, and hacking.  Last time I checked our friends in the three letter agencies are good at all three.  This is why 'trust me security' is fundamentally flawed.