Crypto tracing, attribution evidence, and expert testimony standards
Free. Watch without an account. Sign in to save progress and take the exam.
Up next
ESI discovery, spoliation risks, and arbitration strategies for digital assetsTranscriptThe player shows YouTube’s automatic captions, which are machine captions we do not review. The full transcript is below.
00:00:00Yeah.
00:00:01And I will talk about it a little bit at a, you know, it's sort of a higher level.
00:00:06And I think this this may get us into this.
00:00:10These are points that have been addressed in some of the case law that has come out, too.
00:00:13Right. There is both.
00:00:16There is a deterministic aspect to the blockchain where it just shows it.
00:00:20It will show you for a blockchain like the Bitcoin blockchain, what transactions were executed.
00:00:27What addresses were the sender and recipient and the amounts in some kind of in some kind of timestamp, as we mentioned earlier.
00:00:36So that's that is a fact.
00:00:40Right.
00:00:41I mean, one question at some point, those that type of information might not even require like an expert.
00:00:46I would certainly use an expert these days because it is it is new and technical to many people.
00:00:51But it's actually fairly straightforward.
00:00:53What is more complex in a lot of a lot of crypto tracing tools, chain analysis has been mentioned.
00:01:02T.R.M. Labs has one eclipse.
00:01:05They they do more than that.
00:01:07They do more than just map the actual movements appear on the blockchain.
00:01:12Right. They often do what they still call clustering, which is they identify addresses that likely are likely to share common control.
00:01:19This is this is important and it is it does.
00:01:23It absolutely is probably going to require expert testimony for quite something.
00:01:27I'm not sure what has become common knowledge, because this is based on inferences collected from with information collected from different sources beyond just the bare ledger information that is on the blockchain to from a mountain state.
00:01:38Right.
00:01:39Those are the data points, kinds of data points that is on the blockchain.
00:01:42So you have they use like a variety of different techniques to associate addresses because addresses, if you don't know this, addresses are are sort of a limitless resource.
00:01:53People talk about wallets and addresses.
00:01:56You can create essentially unlimited number of addresses for any purpose.
00:02:01Right. And there's no real cost to doing so.
00:02:04So complex transactions or even kind of simple transactions can actually involve multiple addresses, especially.
00:02:10And if someone is trying to actually cover their tracks, they were almost certain.
00:02:14One of the classic ways to do that is to move money through multiple addresses, mixers and tumblers that can obfuscate the chain.
00:02:23But even if they're not taking any steps to obfuscate their track, their tracks, these transactions can involve multiple addresses and wallets.
00:02:30And so the crypto tracing tools that many of us rely on and that you see where this evidence actually comes from when it's introduced in court usually or used in an investigation does have relies on some of these inferences.
00:02:43And those inferences can be based on if they have information about the IP address, right, that was used to execute the transaction.
00:02:55The connection, the connection between multiple addresses, they're involved in the same transaction and how that pattern plays out.
00:03:04And even from the crypto analysis companies, proprietary information, they blend that information, right?
00:03:11Those crypto tracing tools are used by litigants.
00:03:14But they're actual they're probably the biggest market is AML and transaction monitoring and investigations types functions.
00:03:21So they include information drawn from a non-blockchain sources, too, that can attribute different addresses.
00:03:29I mean, when I was a criminal investigator, we used, you know, one of our key cases, it turned out that one of the individuals of Bitcoin exchanger is it advertised their Bitcoin addresses.
00:03:38So we use that for attribution to help build the chain to like a real world person.
00:03:42But that information can come from offline or off chain, too.
00:03:45So when you are looking at an analytic chain in a in a blockchain tracing tool, it is going to be a blend of hard, immutable ledger facts and inferences and information sources drawn from those other sources.
00:03:58And that's where a lot of the expert testimony is necessary.
00:04:02And it's interesting, actually, sometimes I was involved in a case, randomly a PSLRA securities case where the class action that this lead plaintiff representative of a plaintiff, the class representative had disclosed their crypto address on what they thought were anonymous form
00:04:28where they've made a host of very disturbing kind of comments.
00:04:32And that was uncovered in discovery.
00:04:35I mean, it was anonymous, but we had, you know, the parties had the guy's crypto address and then found use that address to find his post on this form.
00:04:46And that resulted in the disqualification of the class representative.
00:04:49So it's, you know, all kinds of things can come up.
00:04:53There's also the issue of Dauber. I think the slides sort of speaks for itself.
00:05:05So I just think for lawyers, one of the things you should really pay attention for is separating inference from fact in the report because it often gets.
00:05:17I don't want to say lawyers write it, but I would say that sometimes an inclination for legal counsel to conflate inference from fact in the report.
00:05:28And as an arbitrator and also as an expert, I can tell you that it's really important to draw out those differences for the reader and for the prior fact.
00:05:39I know, Judge, you want to weigh in here and then we'll keep it moving just in the interest of time.
00:05:45Absolutely. So I think two points. Number one, I think people, again, think that Daubert is somehow going to save them if they're the opponent.
00:05:54And as the Stirlingov case that we had a slide on earlier shows, courts are going to be fairly liberal in allowing this type of testimony in, or at least the case law is going that way, which I think is exactly what the FRE are designed to do.
00:06:10But that's a separate question from this slide, which is how do you have a credible expert?
00:06:16And I think that sitting as the tribunal a lot of times, the biggest thing that experts do to lose credibility with me is become advocates as opposed to expert witnesses.
00:06:28And when they admit this is what the fact is, this is the inference I'm drawing, they gain a lot more credibility than if they try to pretend that the inference that they're drawing is something that they're establishing with facts.
00:06:41So, Daniel, I love this slide and the fact that you included it.
00:06:47Yeah, so I will take that win and move us right along.
00:06:53For ESI, this is another thing people really struggle on. Smart contracts are coded on the blockchain.
00:07:02So discovery is like a more complicated concept. If you like, give me all the smart contracts, well, you can get them yourself.
Automatically generated; it may contain errors.
Study materials
Notes, flashcards and practice questions for this lesson, reviewed before publication.
Study notes
Deterministic blockchain facts versus inference-based tracing
Justin Herring explains that a blockchain such as the Bitcoin blockchain has a deterministic layer: it shows which transactions were executed, which addresses were sender and recipient, the amounts, and the timestamp. He describes this as a fact, and notes that this type of information might not even require an expert because it is fairly straightforward, though he says he would still use an expert today because the subject is new and technical to many people.
What crypto tracing tools add beyond the raw ledger
Justin Herring notes that commercial crypto tracing tools, such as TRM Labs and other chain analysis tools, do more than map the movements that appear on the blockchain. They also perform clustering, meaning they identify addresses that are likely to share common control. He states this clustering work is likely to require expert testimony because it rests on inferences built from information collected from sources beyond the bare ledger data on the blockchain, and this is not yet common knowledge.
Why addresses are easy to multiply and obscure
Justin Herring points out that addresses are a limitless resource: anyone can create an unlimited number of addresses for any purpose at essentially no cost. As a result, even simple transactions can involve multiple addresses, and someone trying to cover their tracks will often move funds through multiple addresses, mixers, and tumblers to obfuscate the chain. Even absent deliberate obfuscation, transactions can still involve multiple addresses and wallets.
Non-blockchain sources used for attribution
Justin Herring describes how tracing tools blend hard, immutable ledger facts with inferences drawn from other sources, including:
- IP address information tied to execution of a transaction
- Connections between multiple addresses involved in the same transaction and the pattern that creates
- Proprietary information held by the crypto analysis companies themselves
He notes the biggest market for these tools is anti-money-laundering and transaction-monitoring and investigation functions, not litigation, so the tools draw on non-blockchain sources to attribute addresses to real people. As an example, he describes a key case in which a Bitcoin exchanger had advertised its Bitcoin addresses, which were then used for attribution to help build the chain to a real-world person. He states that attribution information can come from offline or off-chain sources as well, so any analytic chain produced by a tracing tool is a blend of immutable ledger facts and inference drawn from these other sources, which is where much of the need for expert testimony arises.
PSLRA case example involving a disclosed crypto address
The panel describes a PSLRA securities class action in which the lead plaintiff, acting as class representative, had disclosed a crypto address on a forum the plaintiff believed was anonymous, where the plaintiff made a number of disturbing comments. The parties had the plaintiff's crypto address and used it to find the plaintiff's posts on that forum during discovery. This was uncovered in discovery and resulted in the disqualification of the class representative.
Separating inference from fact in expert reports
Daniel B. Garrie advises lawyers to pay close attention to separating inference from fact in an expert report, because legal counsel sometimes has an inclination to conflate the two. He states it is important to draw out this difference clearly for the reader and for the finder of fact.
Daubert standard and what makes an expert credible
Hon. Jackson Lucky makes two points on this topic. First, people on the opposing side often think Daubert will save them, but as the Stirlingov case shows, courts are being fairly liberal in allowing this type of testimony in, which he views as consistent with how the Federal Rules of Evidence are designed to work. Second, and separately, he addresses what makes an expert credible to a tribunal. He states that the biggest thing that causes experts to lose credibility with a tribunal is becoming an advocate rather than an expert witness. An expert who openly states "this is the fact, this is the inference I am drawing" gains more credibility than one who tries to present an inference as if it were an established fact.
Discovery challenges for smart contracts as ESI
Daniel B. Garrie turns to smart contracts as a form of electronically stored information. Because smart contracts are coded on the blockchain, discovery becomes a more complicated concept than with ordinary ESI. He notes that simply asking an opposing party to "give me all the smart contracts" is not a meaningful discovery request, because that information can be obtained directly without a request.
Flashcards
13 cards. Test your recall, then flip to check.
Select the card, or press Space, to flip it. Left and right arrow keys move between cards.
1 of 13
Practice quiz
Practice — not graded, not the final exam