Did Loaded (aka “Jimmy Zhong”) taint Greg Maxwell’s wallet?

[nullius]

For my continuing responses Re: Mixer restrictions, I reviewed the start of a classic thread I hadn’t perused in eight or nine years:  gmaxwell’s 13-year-old I taint rich! (Raw txn fun and disrupting 'taint' analysis; >51kBTC linked!).  Woah.  I actually sort of vaguely remembered that a participant described themselves as “loaded”, but I didn’t make the connection that it was the Loaded (u=73652)—who allegedly had an alt mb300sd (u=36986)—and who, much to his dismay, now has a Wikipedia article about him.

Anyone else who wants to be associated with 40k BTC

Code:

0100000003c8e2c94e9

69d9d66aae4812b6cf156f32267b773fb2118db696bb847ebd3454a198b59fbd

I've handled pricy assets before, but perhaps thats the most I've ever had move on a single keypress. Very cool.  I'll have more outputs up in a minute.

LOL, I definitely want to be associated with 40k BTC.  Very cool.  Is the offer still open, Loaded?

...soooo...  Did this actually taint the wallets of Greg Maxwell and other participants in the thread?  Sorry if this was discussed to death while I was sleeping; it’s not coming up on a search.


Notice to moderators:  For the following reasons, I’m posting this on a board where gmaxwell has the mod bit set, and then self-reporting the OP.  I almost posted it to Investigations.

• Courtesy to gmaxwell.  I’m not stirring shit about him.  However, he did start the “I taint rich!” thread to evolve the discourse about taint.  Maybe he got more than he bargained for?
• I’m sort of protesting the dox of Loaded and their putative alt mb300sd.  Just because the U.S. government doxed them and it was headline news, that does not make it compliant with forum rules.  And as we all know, compliance is a buzzword of pure virtue throughout the Web 4.0 fintech industry!  For my part, I think it’s none of anyone’s fucking business what Loaded/mb300sd has printed on their government-issued slave card.  Anyway, I will let the forum’s Rules Compliance Officers (= moderators) check my compliance with the forum rule that restricts doxes to Investigations.

As such, I wouldn’t strenuously object to moving this to Investigations.  However:

• I do think this should be a public discussion, per the purpose of the “I taint rich!” thread.  Maybe gmaxwell got even more of proving his point than he could have wished for?
• I guess that Loaded may have lost access to their forum accounts.  Just in case they ever check the forum to see what people say about them, this thread should be readable 100% anonymously—and it shouldn’t be hidden in a place requiring Full Member status to access.

Not joking about the latter.  The criminal prosecution was a transparent cash-grab that reeks of hypocrisy.  Funds were not restored to the victim of the theft—and if Loaded was a danger to anyone, they’d still be in a cage.  Just to pad their own wallet, the U.S. government used a $5 wrench attack to rob the bitcoins that Loaded stole first from a drug dealer, plus the bitcoins that Loaded earned by honest means.  USG are armed robbers, hardened violent criminals under colour of “law”; that’s worse than a nonviolent hax0r idly jiggling a security flaw in a DNM checkout the DNM needs to hire other hax0rs to secure, as the police don’t honestly care about protecting it.  It would be nice if Loaded gets their life together, and returns to cryptocurrency.

After all, Loaded taught us an excellent lesson about why we need to make Bitcoin fungible.  Without on-chain privacy, every UTXO is an NFT.  It taint right!

Blacklist providers maintain sets of outputs that are blacklisted or tainted in some way, for instance because the owner reported them stolen or because it’s believed they are owned by some criminal enterprise.

If you see this as a good thing, you hate freedom.

[Danish Ali]

What's important to note is that in this instance we are dealing with an instance where multiple parties had contributed to the transaction. Both Loaded's input and gmaxwell's input were consumed in the transaction, with the outputs returning the funds to the respective parties. This means that if you were a blockchain analyst, you could use the common input ownership heuristic to group these two inputs and assume they are linked to one entity.

However, this is not evidence that they are owned by the same party. Bitcoin only ensures that the inputs are authorised to spend their respective UTXO. There is no requirement that all inputs within a transaction are owned by one person. All we know is there's been a public association between them. The original experiment demonstrates exactly how this is a limitation.

[nullius]

The original experiment demonstrates exactly how this is a limitation.

I understand the purpose.  That’s why I love that thread, I have it bookmarked, and I looked back upon it 13 years after the fact.

In retrospect, it succeeded in that purpose much more than intended.  That could be a good thing or the opposite, depending on how one looks at it.  Be careful what you wish for?  If you set up an innocuous public demonstration of why taint is a bad concept, you just may entangle your wallet with the future subject of a Wikipedia page that says, “Known for: Largest theft of Bitcoin in U.S. history”.

It’s surprising that I can’t find any discussion of this from the time period when Loaded was in the news.  More surprising that no one seems interested in it now.  Perhaps I just didn’t locate the thread with the big discussion before?

[Danish Ali]

I understand the purpose.  That’s why I love that thread, I have it bookmarked, and I looked back upon it 13 years after the fact.

In retrospect, it succeeded in that purpose much more than intended.  That could be a good thing or the opposite, depending on how one looks at it.  Be careful what you wish for?  If you set up an innocuous public demonstration of why taint is a bad concept, you just may entangle your wallet with the future subject of a Wikipedia page that says, “Known for: Largest theft of Bitcoin in U.S. history”.

It’s surprising that I can’t find any discussion of this from the time period when Loaded was in the news.  More surprising that no one seems interested in it now.  Perhaps I just didn’t locate the thread with the big discussion before?

You are right. Whilst there was an aim to have a transaction for demonstration purposes the fact remains that there is a permanent link between gmaxwell's wallet and Loaded's wallet on the blockchain (as you said). This is a great example of how taint analysis can be a bad way of finding possible links. In this case it can lead to a false assumption that both parties owned the same funds. It can also show how taint analysis can give false conclusions. There is a historical link between them however there is no evidence they were commonly owned.

[Quickseller]

That thread shows that grouping addresses together after a single transaction with common inputs is insufficient to prove one person owns both addresses. I vaguely remember some kind of service by blockchain.info that would group transactions with ~10 inputs and ~10 outputs with "random" input/output sizes, and would go through x rounds of similar transactions. The idea was you could put in 0.55BTC as an input, that input would be joined with other inputs of some range of sizes, would have several outputs in some range of sizes, then the process would repeat itself until you eventually receive an output of 0.54xxxxxxxxBTC.

If you use blockchain analysis with additional publicly available resources, you can probably group addresses together with decent reliability. Even then, it often will not be necessary to even use that evidence to prove anything. According to a CNBC article linked in the wikipedia article, Loaded showed law enforcement his wallet with the stolen bitcoin. (it was probably also not smart to report the bitcoin, that he had stolen, as stolen).

[nullius]

This is an extraordinary case, of course.  What most concerns me is the ordinary case:  People’s UTXOs are autojudged every day by blackbox robots—with financial and privacy consequences that can be costly, or even catastrophic.

I doubt that Greg Maxwell is at any personal risk.  I presume that the Loaded case was handled by professional detectives, not by stupid rinky-dink cops; if an intelligent human ever investigated wallets linked to Loaded by the common-input-ownership heuristic, I guess they probably would have found the I taint rich! forum thread.  But it would be fascinating to know if those blackbox AML-bots have negatively “risk scored” any of gmaxwell’s coins because of this.  Or for that matter, coins owned by any other participant in that thread.

If you use blockchain analysis with additional publicly available resources, you can probably group addresses together with decent reliability.

What is “decent”?  It is never even close to 100% reliable.  Yet when robots pass judgment, people get shotgun-KYCed, and lose either their funds or their privacy—funds get frozen—all sorts of bad things can happen.

Quickseller, I know you have relevant technical understanding.  I ask this not to be argumentative, but for your opinion:  Even setting aside arguendo the principles of privacy and freedom (which I will never surrender), is there any combination of blockchain analysis heuristics sufficiently reliable, in simple fairness, to support a robotic “AML check” as judge, jury, and executioner, the decisions of which can be appealed to humans only in unpleasant, invasive processes where the coin’s owner bears the burden of proof of innocence?

Overall, I think the concept of coin “taint” needs to die.  Wherefore I call attention to this...

Merited by nullius (10)
Anyone else who wants to be associated with 40k BTC is welcome

Merited by nullius (10)
I've handled pricy assets before, but perhaps thats the most I've ever had move on a single keypress. Very cool.

Known for

Largest theft of Bitcoin in U.S. history

...not to grandstand, and not only as I’m preparing to catch up in the discussion Re: Mixer restrictions.  Almost from the beginning, there’s been a fight over whether Bitcoin is fungible currency, or (anachronistically using the term) a bag of NFTs graded by “taint”.  I believe it is the pivotal issue in Bitcoin’s past, which shall decide Bitcoin’s future.

Forum bug:  I can’t merit anonymous posts from the era when the forum allowed anonymous posts.

If you see this as a good thing, you hate freedom.

[Demian Green]

There seems to be a clear difference between transaction history and ownership of funds. Even when the coins of a known origin take part in a transaction, it doesn’t necessarily mean that all parties whose wallets have been involved become tainted. Bitcoin transactions may include more than one input and produce several outputs, making attribution quite difficult

[Quickseller]

I doubt that Greg Maxwell is at any personal risk.  I presume that the Loaded case was handled by professional detectives, not by stupid rinky-dink cops; if an intelligent human ever investigated wallets linked to Loaded by the common-input-ownership heuristic, I guess they probably would have found the I taint rich! forum thread.  But it would be fascinating to know if those blackbox AML-bots have negatively “risk scored” any of gmaxwell’s coins because of this.  Or for that matter, coins owned by any other participant in that thread.

If you use blockchain analysis with additional publicly available resources, you can probably group addresses together with decent reliability.

What is “decent”?  It is never even close to 100% reliable.  Yet when robots pass judgment, people get shotgun-KYCed, and lose either their funds or their privacy—funds get frozen—all sorts of bad things can happen.

Quickseller, I know you have relevant technical understanding.  I ask this not to be argumentative, but for your opinion:  Even setting aside arguendo the principles of privacy and freedom (which I will never surrender), is there any combination of blockchain analysis heuristics sufficiently reliable, in simple fairness, to support a robotic “AML check” as judge, jury, and executioner, the decisions of which can be appealed to humans only in unpleasant, invasive processes where the coin’s owner bears the burden of proof of innocence?

I guess I don't agree with your premise. In the case of Loaded, the government used blockchain analysis to locate a suspect, but they didn't use the analysis to obtain any private information. Law enforcement used the analysis to decide whose door to knock on, and when he answered, he voluntarily (even if under false pretenses) gave law enforcement a tour of his house, and even went so far to show evidence of his crimes. Bro had something to hide, and pointed at evidence that showed he was guilty to someone he knew was law enforcement. It was basically the opposite of "if you have nothing to hide, why don't you let us look around". There isn't any amount of privacy protections that would have stopped him from eventually getting caught.

Going to the generalized argument, I think you know the saying not your keys, not your money. So I think you know the risks of handing over your money to an entity that doesn't verify your identity ahead of time, but can if they decide they "need" to. Chances are you wont be the first person someone asks for KYC documents from, and you should know this ahead of time, especially if you are handing over large amounts of bitcoin.