'50% of transactions were fraudulent' when Steam accepted Bitcoin for payments, says Gabe Newell | PC Gamer
There was another article somewhere that talked about some of the specifics of what Valve considers fraudulent, but I can't seem to find it again, so likely it was a link from one of the various articles quoting from this one.
Was the fraud related to gamers buying games with crypto and then selling their account for $?
Or was it fake games that were put on the store, in order to use Steam's crypto-checkout feature, which insiders used to launder crypto?
Or?...
'50% of transactions were fraudulent' when Steam accepted Bitcoin for payments | Hacker News (ycombinator.com)
I couldn't find a description of the fraud in this thread. So, still mystified as to the actual fraud that Gabe is referring to.
The first theory there is to do with zero-confirmation fraud. But actual instances of zero-confirmation fraud are hard to find. You have to have been fairly expert with the command line in a bitcoin wallet in order to have performed zero-confirmation fraud in the period 2016/4 to 2017/12 when Steam purchases were possible. I'm certainly not saying it's impossible.
[2016-09-17] Steam Switched Back to Zero-Confirmation Bitcoin Transactions (bitcointalk.org)
The concept of zero-confirmation fraud is very old and BitPay, the service that processed bitcoin payments on behalf of Steam, had monitoring in place to combat this, back then:
How do I prevent double-spend fraud on unconfirmed payments? – BitPay Support
The "Paid" status is the "high risk of double-spend" in the table on that page, corresponding with zero confirmation. Back in November 2016:
BitPay Deploys Advanced Merchant Risk Mitigation for Transactions
Unfortunately in 2016 the BTC network decided to add the replace by fee feature which, in terms of the whitepaper definition of bitcoin is
always fraud. But that's another story.
Replace by Fee allows a transaction to be re-sent with a higher fee. But it's possible that the re-send does not include the address of the merchant that originally invoiced the customer, thus committing fraud against the merchant:
zero confirmation - What does a merchant need to do to reject RBF transactions - Bitcoin Stack Exchange
Anyway, BitPay monitoring can easily see that a transaction was sent with the option to perform replace by fee enabled (the replace by fee transaction would come later). If that option is set to on for the original transaction, then BitPay can require 1 confirmation (as explained in that answer), which then means zero-confirmation fraud isn't possible for the purchase. So using the replace by fee feature of bitcoin to defraud Steam was impossible once BitPay implemented this monitoring. It's unclear when BitPay actually activated replace by fee monitoring though.
Or maybe fraud was actually defined by Steam as accounts that were banned after games were bought (with bitcoin), e.g. for phishing, scamming - or maybe just breaking some rule. "This guy bought loads of stuff with bitcoin. Later we banned him for being an arsehole on the forum. Fucking fraudsters, they all use bitcoin."
So, I cannot determine the actual fraud mechanism, which is why I asked for clarification.
So in the end, it's impressive how Gabe says something almost meaningless and it's lapped up at face value. Transaction fraud would have been extremely rare, so I can't see how payment with bitcoin, per se, was the fraud technique.
Your Bitcoin is no good here—Steam stops accepting cryptocurrency | Ars Technica
No mention of fraud.
