Data Finnovation
@DataFinnovation
Followers
8K
Following
321
Media
883
Statuses
6K
Traditional & Crypto Data CEO @chainargos TradFi and Crypto on one account w/ work done to same standards
Singapore
Joined August 2018
on compliance impossibility:. now that folks are focused on compliance because they are afraid of facilitating terrorism xfers. lets look at how it's not possible for the types of systems people want. @BenCharoenwong posted our preprint last week.
1
3
12
just catching up and the arguments on both sides of this case are a fascinating study in, well, in something.
0
0
0
roman storm's lawyers really argued there's no evidence a user registry smart contract has ever been used? ok. here is a still-actively-used ondo finance contract kycregistry that does exactly that deployed months before storm was arrested. many examples.
etherscan.io
Contract: Verified | Balance: $0 across 0 Chains | Transactions: 19 | As at Aug-22-2025 04:22:15 AM (UTC)
1
0
0
We've said repeatedly: govt should have charged team for the time it was custodial and sorted this out after publishing trial exhibits/detailed plea agreement. Or "someone" could just leak all the privileged comms at the funniest possible time. Max amusement is post verdict fwiw.
0
0
2
Nothing good there yeah. If lawyers signed off they should be disbarred. If fund overruled/ hid from lawyers it's a conspiracy. If the team didnt tell them we have a pvp situation.
1
0
1
Read that post. It says:. Note: in order to aggregate deposits and withdrawal to the Merkle trees, someone needs to run a script called the root-updater. So long as someone out there does this, the system works smoothly and trustlessly.
1
0
0
That was may 2020. dragonfly didnt invest till December. What happened in December? Oh the token launch. Again, teams own words. Presumably the lawyers signed off on this structure.
tornado-cash.medium.com
Tornado.Cash has become the largest privacy solution on Ethereum today. Tornado.Cash is been fully autonomous and decentralized, but it’s…
1
0
0
In may 2020 the tc team publicly admitted they'd been running a centralized custodial platform for some time. @hosseeb lawyers surely knew this im "guessing" unpublished comms describe the old design as clearly outlaw.
tornado-cash.medium.com
We have some great news! As we promised from our previous report, we have set the operator address to…
1
0
0
The dragonfly- tornado cash situation relates to a number of public statements by the tc team at that time about how the protocol really had a few trusted centrlaized components. Some docs that prove the quoted statement is, well, read on. .
Dragonfly invested into PepperSec, Inc., the developers of Tornado Cash, in August of 2020. We made this investment because we believe in the importance of open-source privacy-preserving technology. Prior to our investment, we obtained an outside legal opinion that confirmed that.
1
1
4
What's funny is hyperliquid and thorchain have same faux solution: centralisation. This imposes same liabilities on the doxed founders given the easy access to their code and personal comms. We are still early. Though some may not survive until later. Won't be pouring one out. .
Most high IQs have left the @THORChain community along the way. We are left with a bunch of retards arguing if ILP is a good idea. Hyperliquid did 400m on their spot market today and 9b on perps. They don’t even care/talk about spot markets. AMMs are inefficient and antiquated.
0
0
2
I actually agree with taylor* and @xachxbt so that's got to be right then. *I can't reply because she seems to have blocked me. That says what it means for us to agree. .
@tayvano_ Finally got time to review the txns you shared and I agree. Idk how you mess up the tracing that bad as a firm to where you couldn’t properly follow instant exchange deposits 1 hop from a theft address and then follow subsequent txns down the wrong path to Tornado…. It’s.
0
0
1
What's fascinating here this is the long-discarded yahoo web search style where important sites are curated vs Google scheme where rankings emerge from network. If genai summarises well this works as a cheap spokesperson but it can't discover anything new (see yahoo vs Google).
The new Grok genuinely runs a search for "from:elonmusk (Israel OR Palestine OR Hamas OR Gaza)" when asked "Who do you support in the Israel vs Palestine conflict. One word answer only."
0
0
1
It's amazing people think quickly resolving one sentence wagers is a tech problem. It's almost like the designers and users of these systems have no experience of the real world. This is why trading floor bets usually fall apart in docs disputes, among many examples.
hey @Polymarket bros . what’s going on here? shouldn’t this be resolved to yes already?. Even the guy that created the suit said it’s a suit, your own team is calling it a suit. I don’t think it gets any clearer than that, but UMA whales are manipulating it anyways (again)
1
0
6
Eventually courts will opine on what are really just the values of r and s the us legal system expects. There will eventually be breathless reporting about ridiculous data sources because a) it's software and b) it's govt databases. It's debugging & data science in real time!.
0
0
0
The govt falling to look at prison records is clearly incompetence. Not even trying to check super-reliable data the govt already has is bad. 0 clue if that was done before so 0 opinion but to note checks for federal prisons must have been done as news on such deports long common.
1
0
0
If the app uses, say, chuck-e-cheese membership application data you need more process. Thats dictated by legal principles around r and s. Nobody expects 0. It's software.
1
0
0
Is it taking names off utility bills or sources with just an initial and surname? Those would give a higher r. And you'd expect the process to give s<r fixing these sorts of expected errors. The legal system gives principles for acceptable r and s values.
1
0
0
Now r is positive for a few reasons. First it's software. Second it's gonna rely on imperfect govt data. If it's pulling, say, prisoner records you'd expect v few mistakes. Drivers licenses w/ dob also probably good.
1
0
0
The app has a false positive rate. That's how often it flags someone as illegal when they are legal. Call that r. Then there's a process that's followed with a false +ve rate post process. Call that s. If there is no real process r=s. You'd hope s<r.
1
0
0
This is fascinating because it's fundamentally a statistics question we can model v easily with unknowns we will be able to measure fairly reliably in the near future. .
Under DOGE, there was an app developed called Alien Tracker or Atrac. Basically, it's early stage but it tells you WHERE these people with deportation orders are. It also has a screening of people with criminal convictions. And it has 700,000 of such people not from ICE database.
2
0
1
It's f university you're supposed to be learning. Learning that is a big part of life. High school baseball players pk fair enough. Also don't abuse the learning disabled. Fine.
0
0
0