Topic: IBM Find Only 2% Of Bitcoin Transactions Are Unlwaful - page 2. (Read 484 times)

The fact alone that this kind of research can be done easily and publicly is a huge factor for bitcoin legitimacy . And this is the thing that most people do not understand, everything is public, all the transactions. In contrast it is very hard to map bank transfers.

In fiat economy the share of shadow economy is 15-40% and sometimes even more, I think thinking that only 2% of Bitcoin transactions are unlawful is quite naive. You don't need to sell drugs or weapons to make illegal transactions, simply working or running a business without a license already counts as unlawful, I believe that trading is the biggest use case of Bitcoin at the moment, but after that there's a huge amount of services that very likely belong to shadow economy.

However we get to know good about cryptocurrency usages there is negative news regarding the same. This can't be stopped, the change can't be experienced overnight. The change happens slowly and what we have with this analysis about the unlawful transaction between bitcoin and fiat is the reality which people find it hard to believe.

It's great that we are trying to use AI more, but this study does not seem to be talking in favor of it. The catastrophic part is that the vast majority of transactions (77%) were unclassified. This means we don't know the percentage of legal transactions to illegal ones even roughly, and it can be that Bitcoin is largely used for illegal stuff. And the very fact that we cannot be sure shows why the authorities might not be happy with it.
On the other hand, what kind of illegal are we talking about here? 'Cause financing terrorism is one thing and not paying taxes because the system in your country is confusing is another.

I figure the goal was to classify each TX as either unlawful or legal. Alas, only 23% managed to get one of those tags, whilst 77% were left unclassified, meaning, I figure that they could not trace nor tie these to either unlawful or legal senders/receivers.

If the wording in the article is as intended (and I’ve checked multiple sources and found nothing better), it means that the phrasing should state that “at least 2% were deemed unlawful”, since there is an immense unclassified 77% of the TXs in the study sample.

We could play the devils advocate here, and just play around with the classified TXs. That would lead to saying that the classified sample group is of 46.000 TX (2%+21% of 200.000). Out of this classified sample universe, 8,70% would be unlawful (4.000 TXs / 46.000 TXs) and 91,30% legal (42.000 TXs out of 46.000 TXs). I don’t particularly like this focus, but it goes to show that the unclassified group is important in size, enough to make the numbers shift wildly from the “only 2%” range provided by the title of the article.

Note: It would also be interesting to know how the 200K TX sample was selected, in order for it to be representative and sparce. For comparison reasons, the current number of BTC TXs/day is around 348K, and the historical total of nearly 444M.

I wonder what parameters have they used in order to come up with such a conclusion. Don't get me wrong, AI is a really interesting and useful technology although sometimes I doubt whether what it spews out are really of factual nature. For my understanding, the AIs of today are still largely controlled by its human engineers, and algorithms are also constructed by humans, so the results can still be biased in some way though again, there's not much discussion regarding the parameters used in the article. 77% of unclassified data is still a huge margin of error to conclude that only 2% are really illicit transactions, and so the research may stumble upon something but for now, it's not really that useful to say the least.

Any transaction may have unlawful things not only with bitcoin and other cryptocurrencies there are sure unlawful things that need to be confirmed under fiat money as well But Any kind of currency has unlawful things that haven't been check by the radar of the authority or government, and one thing is certain criminals would surely want a payment method that is off track the radar and Bitcoin and other cryptocurrencies is best fit that category because of its anonymity but as you have said there are just 2% tracked bitcoin transaction that is unlawful, well maybe the 77 classified ones were pretty certain to have things more anonymous but we can not know for sure.

Interestingly they do not do this kind of research for fiat currency. It's like unlawful act never happened with fiat.

How about the coins those were hacked? They should fall in the category of unlawful.

2% is still a lot, i guess. It will be intresting too know % of unlawful fiat transactions

Maybe you're talking about this one: https://info.elliptic.co/whitepaper-fdd-bitcoin-laundering

This study (from the same people, Elliptic, who helped IBM in the new one) was released in 2018 but it took place from 2013-2016 and it pretty much says that, in those 3 years, less than 1% of btc transactions were coming from illicit activities. Most of the volume from the illicit transactions came from Europe, and the main source for all the illicit transactions was (and probably still is) darknet maketplaces.

This isn't the first such publication by Elliptic. Can't really be bothered to dig up older ones but I can almost swear there is at least 1 in 2016 that mentions an even lower number (1%). Might have been another study from another firm but the methodologies have likely improved.

2016 was already when they noted the trend moving to other coins, this was just fresh after Silk Road.

To note, unlawful doesn't necessarily mean illicit.

it's 77% you made a typo, I feel like they rushed to publish the result even if it is not conclusive. 77% unclassified transaction is too high to claim that the study is a success.

There was nothing in the article in terms of an explanation on how they arrived at the results. There should have been some information regarding the methodology used for classification.
What are the 71% unclassified transactions? When those many are unclassified, how can they claim that "only 2% are unlawful".

There is additional information available regarding the methodology on the website of Elliptic. Seems like they use their "forensic tool" to trace origin of bitcoin transactions with a proprietary set of addresses whose owners are known to be associated with dirty bitcoin. Dirty bitcoin refers to those addresses that were at any time associated with the Darknet marketplaces of yore.

Interesting!
I suspect a large % of fiat transactions are unlawful in one way or another, ranging from tax avoidance, money laundering, drugs etc, pretty much the same with any means of payment. People do illegal shit & that’s that.

A London based company called Elliptic recently made a claim that only 2% of Bitcoin transactions are unlawful. To respond to that a partnered study was created by IBM and MIT to check whether the statement was actually true. It "utilized Artificial Intelligence and deep learning algorithms to reach its conclusions, represents something of a landmark for the use of AI in the blockchain and cryptocurrency industry. Of the 200,000 transactions analysed throughout the study, two percent were deemed unlawful, 21 percent were considered legal, while the remaining 77 percent were left unclassified. A key element in the success of the recent study is the utilization of artificial intelligence throughout the procedure. The integration of AI into the process of AML(Anti Money Laundering) protocols is significant in that it highlights how far the technology has come. Artificial intelligence is now sophisticated enough to both analyse and navigate the complexities that the blockchain presents.
Artificial intelligence’s role in the world of cryptocurrency and the prevention of criminal behaviour is already fairly well defined at this point, but the integration of the technology into the blockchain industry could be a game-changer of significant proportions." Especially considering the reputation Bitcoin is creating for itself for unlawful and illegal activities despite the evidence this research has created against it. "Today there is still a lot of skepticism of the blockchain and crypto worlds, more so the latter, but that is to be expected of an industry that is still relatively young and finding its feet. What cannot be denied is the vast potential that blockchain technology offers, and the integration of AI is yet another facet of this industry that could change the way we go about business in the future.
The Elliptic report isn’t the first time we’ve seen AI technology used in the blockchain and crypto world. Companies such as Singapore-based DeepBrain Chain and Velas, a startup based out of Europe have been forging a path for quite some time now.
For Alex Alexandrov, CEO of both Velas and crypto payments gateway CoinPayments, the role of artificial intelligence in the blockchain industry is simple: To solve problems. “Here at Velas, our purpose is to address and fix existing issues and challenges faced by most existing Blockchains, such as centralization for example, or 51% attack, nothing at stake problem, scalability, security, high upfront expenses and so on. This is done by using neural networks optimized by artificial intelligence to enhance its consensus algorithm.”
In a recent interview with a leading cryptocurrency news outlet, DeepBrain Chain CEO He Yong explained that DeepBrain Chain aimed to “connect all the idle AI computing resources globally so that the entire network can share computing power.”
We are seeing the technology gain more and more acceptance from academics and developers alike, which can only be a good thing for overall progress within the space. The work carried out thus far by companies like Velas and DeepBrain Chain is shining examples of what is possible when AI is utilized alongside blockchain technology."
-Source https://www.investing.com/analysis/elliptic-data-ibm-find-only-2-of-bitcoin-transactions-are-unlawful-200453542