> The attacker used social engineering to induce Drift Security Council multisig signers into pre-signing transactions that appeared routine but carried hidden authorisations.
So much for the "Security Council". What an embarrassment to be in a team/org like that and fail your most basic duty which would be "look at what you sign".
That was inevitable, and all designs like that will eventually yield the same outcome.
The people who should be embarrassed are the ones who thought having a group of humans routinely review (possibly complex) transactions for correctness, with no ability to undo/revert the outcome, was a good idea.
This is conveniently suspect, no? “Drift migrated its Security Council on March 27 to a new 2-of-5 threshold with zero timelock. That eliminated the delay that would have allowed detection before admin actions took effect.” This was after the perp started working on the heist earlier in the month.
The multisig UI/UX is a real and long term difficulty for any governance council. "Please sign this opaque transaction with binary data, it represents our agreement. I promise." For a while maybe ten years ago I worked with MakerDAO on this problem - at the time the idea was a separate auditor for proposed transactions.
This general attack pattern is: get a lender with good collateral to call your bad collateral good, then swap collaterals, and it's a known and bad attack vector; the ongoing tension between innovation / speed and caution continues.
There's probably a flash-loan multiplier angle here for an even worse attack; I'm imagining chaining a liquidity change in the trusted price oracle for the CVT token in the middle of the swapping. Anyway, upshot - don't loan against North Korean attack tokens. Put it on the list.
Back at the top of the crypto hype cycle I wouldn't be surprised to see a project survive even a situation like this one, but now that the hype has died down is it still possible to come back from a loss of this magnitude?
this is a beautiful attack, the way that multisig signers were compromised with innocuous signatures in advance, without really compromising private keys
from the pre-funding to a virgin address, to the bundler, to the exit strategy to decentralized assets
to the protocols exposed but functioning perfectly under the stress test - props to Jupiter! - and the optional insurance protocols functioning decently, all while people point fingers at Circle for their bridge working perfectly, it's not even clear what people want them to do specifically! All of these aspects of web3 are working great, and it's easy for a cynic that only sees these headlines to miss that
It's always entertaining to see worthless idiots lose money on an obvious scam like cryptocurrency. Ha ha. Although in this case it seems that North Koreans might have ended up with actual valuable fiat currency, which is unfortunate.
32 comments
> The attacker used social engineering to induce Drift Security Council multisig signers into pre-signing transactions that appeared routine but carried hidden authorisations.
So much for the "Security Council". What an embarrassment to be in a team/org like that and fail your most basic duty which would be "look at what you sign".
The people who should be embarrassed are the ones who thought having a group of humans routinely review (possibly complex) transactions for correctness, with no ability to undo/revert the outcome, was a good idea.
This general attack pattern is: get a lender with good collateral to call your bad collateral good, then swap collaterals, and it's a known and bad attack vector; the ongoing tension between innovation / speed and caution continues.
There's probably a flash-loan multiplier angle here for an even worse attack; I'm imagining chaining a liquidity change in the trusted price oracle for the CVT token in the middle of the swapping. Anyway, upshot - don't loan against North Korean attack tokens. Put it on the list.
> The funds were used to deploy CarbonVote Token (CVT), a completely fictitious asset
Crypto calling out other cryptos, made me giggle
Back at the top of the crypto hype cycle I wouldn't be surprised to see a project survive even a situation like this one, but now that the hype has died down is it still possible to come back from a loss of this magnitude?
from the pre-funding to a virgin address, to the bundler, to the exit strategy to decentralized assets
to the protocols exposed but functioning perfectly under the stress test - props to Jupiter! - and the optional insurance protocols functioning decently, all while people point fingers at Circle for their bridge working perfectly, it's not even clear what people want them to do specifically! All of these aspects of web3 are working great, and it's easy for a cynic that only sees these headlines to miss that
inspirational, great place to build
But all these new protocols want to do stuff at the expense of trustlesssness.