Bitcoin Forum
September 27, 2026, 09:37:30 PM *
News: Latest Bitcoin Core release: 31.1 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 [40]
  Print  
Author Topic: Large-scale Coldcard compromise (1596 BTC stolen so far)  (Read 13530 times)
Leon Zimmer
Newbie
*
Offline

Activity: 28
Merit: 29


View Profile
Today at 07:48:20 AM
 #781


ColdCard was NOT open source. ColdCard was Source Verifiable.

What's the difference? Open source means anybody can use the code, even in their own work.

Source verifiable means anyone can read the code, but they can't use it in their own work. So, what's the incentive for experts to read it? And even for those who do read it... some bugs aren't likely to be spotted on sight. It isn't until you tinker with it that you realize something behaves unexpectedly.

ColdCard got greedy. They didn't want people using their code. Their hubris and greed led to them getting wrecked, and that would be great if ColdCard users didn't get wrecked too.

Which brings us to Coinkite’s “open” code. The last part is key. If you’d used MIT or GPL, others could fork it or provide fixes for a fee or sell competing products. With the current license anyone can read it, but they can’t make competing commercial products. That means the only people with an incentive to deeply analyze the code are researchers looking to make a reputation, who are unlikely to closely review entropy generation, or people looking to exploit it, who are likely to closely review entropy generation.

Deeply analyzing entropy generation happens if you get paid to do so, or if you’re trying to exploit it. That’s why this bug sat around for 5 years. Coinkite gets the PR of open source, with the legal protections of closed. They got the marketing benefit of both models and the security benefit of neither. Illusory security through questionable licensing.
Cricktor
Legendary
*
Offline

Activity: 1624
Merit: 4492



View Profile
Today at 01:39:47 PM
 #782

Who has an incentive to look closely at the code? At least those, who want to benefit from potential exploits, as happened in the recent Coldcard debacle.

Do security researchers have an incentive? It depends on minimum two prerequisites:
a) a fair bug bounty program
b) a company which is open to a constructive discourse about bugs, issues, flaws of their product

To be fair, I don't know if Coinkite has a bug bounty program in place. Dealing with reported issues, maybe also including bugs, my reception isn't really positive on how Coinkite and nvK handled it in the past (I'm certainly somewhat more negatively biased here, as I paid more attention to negative behavior than the opposite).

This likely creates an environment which isn't quite encouraging for security researchers. I mean, those commonly preinvest their time and therefore money to find something that will or rather should pay off later. If you're in that business, you don't really want to be brushed off by some arrogant dudes who think their product is superior.

Well, and voluntarily auditing the code? I doubt there are many with good enough expertise to do it for free or "for the merits". Has Coinkite ever acknowledged external contributions to improve their code? Genuine question, because I don't know and never cared to pay attention to it because the company management's behavior to date has completely repelled me.

Danish Ali
Member
**
Offline

Activity: 70
Merit: 155


View Profile
Today at 04:38:13 PM
Merited by ryzaadit (1)
 #783

Who has an incentive to look closely at the code? At least those, who want to benefit from potential exploits, as happened in the recent Coldcard debacle.

This is also, I think, a good example of how just because it's reviewable doesn't mean it will be reviewed properly, as the bug was discovered by gmaxwell looking at code. Several people reached the same conclusion by disassembling the firmware. Either way, people were able to go and look at the code/firmware, but this doesn't mean that most people can/will. It takes a highly-skilled analyst to recognize what went wrong here. Openness alone is no panacea, since it took someone as good as gmaxwell to recognize a #define/#ifdef mixup.


This likely creates an environment which isn't quite encouraging for security researchers. I mean, those commonly preinvest their time and therefore money to find something that will or rather should pay off later.

And the real crux: assuming a researcher is able to find such a subtle vulnerability, and finding an RNG whitening bug is no cake-walk either, they may very well decide to exploit it, sell it or report it. Selling/exploitation will get them way more money than bug bounty will. And potentially millions.

So, unless you want to compete with black-market in offering bounties , your company’s bounty policy, however generous, will be irrelevant. Ethics and professional reputation are the only things keeping such researchers from exploiting vulnerabilities they discover.


Has Coinkite ever acknowledged external contributions to improve their code? Genuine question, because I don't know and never cared to pay attention to it

Aside from the RNG issue itself, I think it's important to comment on the treatment of outside researchers. It doesn't take long before it becomes common knowledge in security circles if a company is going to ignore outside researchers. The next person who finds something really serious at this company might not be jumping through hoops to report it for free. At best, they'll say nothing. At worst, they'll sell it on the black market.
ryzaadit
Legendary
*
Offline

Activity: 3332
Merit: 1542


View Profile WWW
Today at 08:08:28 PM
 #784

To be fair, I don't know if Coinkite has a bug bounty program in place. Dealing with reported issues, maybe also including bugs, my reception isn't really positive on how Coinkite and nvK handled it in the past (I'm certainly somewhat more negatively biased here, as I paid more attention to negative behavior than the opposite).

This likely creates an environment which isn't quite encouraging for security researchers. I mean, those commonly preinvest their time and therefore money to find something that will or rather should pay off later. If you're in that business, you don't really want to be brushed off by some arrogant dudes who think their product is superior.

Well, and voluntarily auditing the code? I doubt there are many with good enough expertise to do it for free or "for the merits". Has Coinkite ever acknowledged external contributions to improve their code? Genuine question, because I don't know and never cared to pay attention to it because the company management's behavior to date has completely repelled me.
No wonder you have a negative bias against CoinKite/nVk. I think other people feel the same way you do. Since we all know nVk has been known for anti-open source and has a history of sending someone cease-and-desist letters. In the past few days and weeks, they have also started to block everyone on X. I don't think that's a good reason to block everyone while they should take any word (including a negative opinion) since they already make people lose their entire life-saving.


Source: https://x.com/asanoha_gold/status/2102631552039727456/



This is someone who reported being drained from mk4 even before the tragedy, resulting being blocked from the channels.


I don't think voluntary auditors, researchers, white-hats, and bug hunters want to dealing with these behaviors. Mostly, they're just being ignored.
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 [40]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!