Live data from Hacker News

'50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

pcgamer.com

211–220 of 333 posts

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#211
post #188

Earlier quoted context omitted.

>They wouldn't be blacklists. You only learn after the fact so its more about the analytics. I mean, that goes back to my original question, why would they do this? Does bestbuy run analytics on the cash they receive to see how much % of their cash purchases were "fraudulent"? >Well that's nice but if you get paid with stolen money, you don't just get to keep the money and tell them to fuck off. Steam would be liable…

> Does bestbuy run analytics on the cash they receive to see how much % of their cash purchases were "fraudulent"? Credit card companies do, yes.

Again, that's because credit card companies and/or merchants eat the cost of fraud. That's not the case with cash. The FBI isn't going to be raiding bestbuy's stores looking for stolen bills.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#212

Earlier quoted context omitted.

I'd like to know what Gabe meant by fraudulent too. The whole idea of crypto is that if you send it and it's confirmed you can't get it back, which makes fraud (on the payers side) nearly impossible. You brought up the point about zero confirmations. Zero-conf itself is fairly reliable, it certainly doesn't account for 50% transactions being fraudulent. Bitcoin introduced replace by fee in 2016. In 2017, bitcoin tran…

> The whole idea of crypto is that if you send it and it's confirmed you can't get it back, which makes fraud (on the payers side) nearly impossible. I consider "obtained somebody's credentials or private key through illegal means and then used that to purchase a good with stolen BTC" to be fraud. That the network considers this to be a normal transaction is not relevant.

I wonder, is there any legal precedent for that? Analogy with other transactions could support either interpretation. (Spending stolen cash is not fraud, but using a stolen credit card is.)

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#213
post #177
post #175

Earlier quoted context omitted.

> even if there's nothing untoward about the purchase itself other than the payment instrument. When you make a purchase with a credit card, you claim that you are an authorized user of the card, and that you will pay the bill in accordance with the account agreement. If that's not true, it's fraud. You've knowingly made an untrue claim in order to receive something of value.

From the other comments on this thread, it seems like Steam was using a zero-confirmation payment validator of some kind. That makes bitcoin payments Steam accepted effectively a form of check, and if customers exploited this to double-spend bitcoin, then they paid steam with a hot check. Hence, fraud. Should Steam used a different form of validation? Yeah, that would have let them reject some (most? all?) of these f…

I thought it was always really big news whenever someone even attempted a double spend, regardless of success? So if that’s what he meant, we should have been seeing those cases in the news back then.

Edit: I guess not? https://news.ycombinator.com/item?id=30480133

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#214
post #63

Earlier quoted context omitted.

It provided a great way to buy weed before it was legalized and transaction fees skyrocketed. Bitcoin at $5,000 a coin was extremely useful for transactions. Bitcoin at $50,000 make it less useful because of scaling issues.

You currently pay about 10 pennies for a transaction which is included in the next block. Using Lightning is even cheaper.

Really, googling around, average transaction fee looks about $1.70? Which is much more than I pay through any other service.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#215
post #154

Earlier quoted context omitted.

> They are irreversible and if the Bitcoin is stolen that's not the merchant's problem. This seems consistent with the bitcoin ethos but incompatible how the economy and legal system work. If someone buys a physical item from your physical store with a hot check, a stolen credit card, or marked bills that were just stolen from a bank, you don't get to keep the money, even if police aren't able to recover the item you…

I do believe that you get to keep cash. I don't think I've ever heard of an instance of "dirty cash" getting seized back from a merchant. Do you have an example of that?

https://en.m.wikipedia.org/wiki/Civil_forfeiture_in_the_Unit...

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#216

Earlier quoted context omitted.

Instantaneous payments exist, I use them several times a day, with traditional banking. I never quite understood what problem bitcoin was supposed to solve beyond anonymity.

These are instant payments, not instant settlement. Your bank abstracts the underlying inter bank settlement complexity and aggregates all the minor transactions and transfers that individuals do to a small number of larger settlement transactions, likely on the SWIFT network. When you make a payment, that transfer of money isn't instantly settled. The instantaneous feeling is a service that the bank provides you, bu…

At a high level this comment is correct, but SWIFT is not a settlement network. A message on the network does not actually represent the movement of money between accounts.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#217
post #84
post #43

Earlier quoted context omitted.

Yes they were using the BitPay provider when they were accepting Bitcoin, I assume they were made aware of that it was possible to track the transactions back to fraudulent credit card transactions, which I guess in some way could still make them liable for it.

This is absolutely not what happened. Why come up with such crazy theories when this just comes down to double spends of unconfirmed transactions?

> Another thing was that the vast majority of those transactions, for whatever reason, were fraudulent, where people were repudiating transactions or using illegal sources of funds and things like that.

Seems like it was multiple things and "fraud" is just a catch-all term and Gabe didn't bother to list everything that they considered to be fraud.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#218
post #6

Earlier quoted context omitted.

> * did someone try to send something other then bitcoin to a ₿ address? That would not even register right? Different coins are completely separate and it wouldn't even have registered anywhere and show up in the statistic. As far as I know the addresses are also distinct between coins (so it's not even physically possible) but there are so many now and I've been out of that world for a few years now so that might h…

If it was double spend, I imagine we would have heard the news. It's probably illegitimate money.

Bitcoin's protocol doesn't prevent a double-spend for a 'short' amount of time after payment, it just guarantees that, eventually*, exactly one of those spends would be seen and counted as valid by all compliant nodes (as long as at least 51% of the network is compliant nodes).

But nothing prevents a minority attacker from branching the chain with their own alternate history at any point, although if they did it to a 'historical' block - one deep behind in the chain - then they're doomed from the start, they could never build an alternate chain longer with the main network racing them. Their only hope is waiting for a spend to become in a valid block then immediately start another branch with a double spend, hoping they could convince enough of the network to accept their version of the blockchain where the second double spend is actually the legit one. This gets harder and harder the more blocks build on the main branch, if 5 or 6 blocks build on a block, it's pretty much impossible for all intents and purposes to invalidate that block. (Unless, again, the attacker cares enough to secure computing firepower greater than or equal to 51% of the overall network)

*: "Eventually" is a probalistic statment, theoritically an alien super computer with unimaginable power could rewrite the whole blockchain from 2009 till now in the time it takes one of ours to boot, the blockchain protocol constantly adjusts a parameter that makes mining a valid block always takes, on average, 10 minutes. Assuming no sudden computing advances, this will always compensates for increases in gradual computing power such that you always need the majority's support for dictating what's the longest branch.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#219
post #186

Earlier quoted context omitted.

Shouldn’t it make most sense in contexts exactly like Steam, where the merchant can take the goods back in case of a fraudulent transaction?

It did, and they did. That's probably exactly why they allowed it like that in the first place - if the transaction is fraudulent then simply remove the user's entitlement, but offer instant access for better experience.

“Fraud” and “simply” don’t belong in the same sentence. This is coming from someone who’s spent years working with many fraud departments in major companies. Wack-a-mole doesn’t work, they just scale their account creation.

Re: '50% of transactions were fraudulent' when Steam accepted Bitcoin for payments

#220

Context on what I believe they mean by "fraudulent". Steam back in the day used to accept 0 confirmation bitcoin spends. This means that the transaction has been gossiped on the bitcoin p2p network but had not yet been mined into a block and thus had minimal finality guarantees. Steam could see that they were going to receive a bitcoin payment (when the transaction was mined into a block) and would credit the users a…

You don't even need to do anything on bitcoin side if you value the 0 confirmation. You offer the goods or services before you receive the payment. It's like a restaurant that serve the food first and receive payment later. The restaurant do that because they trust the customer to pay up.
Post reply on HN