Large-scale Coldcard compromise (1596 BTC stolen so far)

[LoyceV]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

[Myleschetty]

I think he believes that all of CEX delisted it at that time but Monero is still listed on CEX like Mexc.
We can not blame those who believe that the ColdCard hack was an insider job because the security flaw was discovered years ago and the ColdCard team did nothing about it.

As a result, people will likely assume that it was a planned attack because no rational person would do that.

The actual red flag was the ColdCard CTO made an anonymous account in Github and started talking to himself through his real account. Why would he do that if there wasn't any deception, and if there wasn't any nefarious motivation.

Which also makes me think about Luke Dash Jr. - Something nefarious is also being hypothesized in BIP-100/FilterBoiCoin. People are starting to say Luke Dash Jr. is a bad actor.

Luke Jr. is always viewed as a mixed bag, and occasionally he will say something that gives the appearance that he is living in a different world.
I read that the ColdCard CTO is the author of the "libngu library that generated the weak seeds of Coinkite based on GPG signatures from his personal key on commits made under both names, but I dont know he pulled such a stunt on GitHub.

https://www.cryptopolitan.com/coinkite-cto-peter-gray-linked-coldcard-hack/

[tvbcof]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

I dunno.  The hardware is still, in my opinion and with limited analysis, the 'best in class'.  In my (somewhat limited) view, they had a lot of terrific ideas about security, and a knack for making the UI quite intuitive.  The documentation is fantastic, and the source code is available.  I just wouldn't be super confident that the binary actually represents the totality of the source code and it could be the case that they have some 'special patches'...I doubt that very many people availed of the feature of being able to build it oneself vs. just scoring one of Coinkit's release binaries.

Actually, I had some suspicions based on the observation that the product was almost 'too good.'  At least based on the common understanding of the size of their development team.

After a thorough analysis of their hardware, some care in sourcing, and a well structured and well know open source development team, the kit would be pretty compelling to me.  At least the Q.

[Meuserna]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

I dunno.  The hardware is still, in my opinion and with limited analysis, the 'best in class'.  In my (somewhat limited) view, they had a lot of terrific ideas about security, and a knack for making the UI quite intuitive.  The documentation is fantastic, and the source code is available.  I just wouldn't be super confident that the binary actually represents the totality of the source code and it could be the case that they have some 'special patches'...I doubt that very many people availed of the feature of being able to build it oneself vs. just scoring one of Coinkit's release binaries.

Actually, I had some suspicions based on the observation that the product was almost 'too good.'  At least based on the common understanding of the size of their development team.

After a thorough analysis of their hardware, some care in sourcing, and a well structured and well know open source development team, the kit would be pretty compelling to me.  At least the Q.

ColdCard just caused the most catastrophic loss in the entire history of Bitcoin hardware wallets due to shockingly shoddy code, and the firmware fix for that caused some people's devices to brick, and the CEO spent the last 2 weeks deleting old posts on twitter as the company basically went into radio silence.... but other than that, Mrs. Lincoln, how was the play?

There's nothing special about ColdCard's hardware. It's cheap plastic. And some aspects of the design weren't even thought through, like the fact that the camera for the QR reader faces up, so you can't even see the screen while you try to scan a QR code. What the whut?

If you already own one and you want to accept the risks that come with staying with Coinkite... do what you think is best. Self custody comes with self-responsibility, so be honest with yourself about the risks. I have trouble imagining that company will even exist in a year, and there's no way I'd ever trust them. I wouldn't trust their hardware. I wouldn't trust their code. I wouldn't trust anything they say.

[tvbcof]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

I dunno.  The hardware is still, in my opinion and with limited analysis, the 'best in class'.  In my (somewhat limited) view, they had a lot of terrific ideas about security, and a knack for making the UI quite intuitive.  The documentation is fantastic, and the source code is available.  I just wouldn't be super confident that the binary actually represents the totality of the source code and it could be the case that they have some 'special patches'...I doubt that very many people availed of the feature of being able to build it oneself vs. just scoring one of Coinkit's release binaries.

Actually, I had some suspicions based on the observation that the product was almost 'too good.'  At least based on the common understanding of the size of their development team.

After a thorough analysis of their hardware, some care in sourcing, and a well structured and well know open source development team, the kit would be pretty compelling to me.  At least the Q.

ColdCard just caused the most catastrophic loss in the entire history of Bitcoin hardware wallets due to shockingly shoddy code, and the firmware fix for that caused some people's devices to brick, and the CEO spent the last 2 weeks deleting old posts on twitter as the company basically went into radio silence.... but other than that, Mrs. Lincoln, how was the play?

There's nothing special about ColdCard's hardware. It's cheap plastic. And some aspects of the design weren't even thought through, like the fact that the camera for the QR reader faces up, so you can't even see the screen while you try to scan a QR code. What the whut?

If you already own one and you want to accept the risks that come with staying with Coinkite... do what you think is best. Self custody comes with self-responsibility, so be honest with yourself about the risks. I have trouble imagining that company will even exist in a year, and there's no way I'd ever trust them. I wouldn't trust their hardware. I wouldn't trust their code. I wouldn't trust anything they say.

I'm not going to spend a lot of time working with you on reading comprehension skills and technical matters, but I will mention that when faced with the problem of not seeing the screen, try rotating the device 180 degrees about the relevant axis.

[joker_josue]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I got curious and went to see. I didn't think it was that big of a discount, not even 15%.

A big discount would be over 50%... even then I doubt they would make any sales. I mean, the worst part is that they're likely to continue making sales, because many people didn't even realize what happened, and how serious it was/is.

[tvbcof]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I got curious and went to see. I didn't think it was that big of a discount, not even 15%.

A big discount would be over 50%... even then I doubt they would make any sales. I mean, the worst part is that they're likely to continue making sales, because many people didn't even realize what happened, and how serious it was/is.

I would estimate that some of the sales have to do with the need for hardware to analyze, and also to look for supply chain attacks.  Probably these would not be through Coinkite though. 
Probably off-the-shelf gear.

I would hope that retail sellers who have stock would hang on to it, and would take note of attempts to either purchase it, or attempts by Coinkite (or suspected associates) to have it returned.

[taufik123]

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

I got curious and went to see. I didn't think it was that big of a discount, not even 15%.
A big discount would be over 50%... even then I doubt they would make any sales. I mean, the worst part is that they're likely to continue making sales, because many people didn't even realize what happened, and how serious it was/is.

yes that's right about the amount of the discount, actually it's too small, it's only 10%, I'm just giving a sarcasm when A huge attack that removed thousands of Bitcoins from users' wallets,
they are still selling and only giving a 10% discount on devices that weren't even purchased when people found out about this hacking incident.
They sell products with failed firmware and people are already very skeptical about using them.

-snip-
I would hope that retail sellers who have stock would hang on to it, and would take note of attempts to either purchase it, or attempts by Coinkite (or suspected associates) to have it returned.

It should be refundable, because of course it is included in the warranty or official partnership has the right to return defective goods caused by the developer himself.

[tvbcof]

...

-snip-
I would hope that retail sellers who have stock would hang on to it, and would take note of attempts to either purchase it, or attempts by Coinkite (or suspected associates) to have it returned.

It should be refundable, because of course it is included in the warranty or official partnership has the right to return defective goods caused by the developer himself.

I'm saying that there may be something quite a bit deeper going on than a couple of thousand BTC being purloined.  At the very least the devices might be useful as evidence and would help paint a picture of their construction and distribution.

If the currently consensus story about a pre-processor macro being mis-used (deliberately or otherwise) is actually true, then the hardware itself may be worth something someday.  Making fairly safe original use of it would involve determining the purity of the hardware component supply chain, and compilation of the firmware by trusted parties (including oneself.)

If there are dedicated efforts to police up devices at the distributor or consumer level, that would be interesting to know of.

[philipma1957]

People are now experimenting with various ways to generate seeds. I wouldn't recommend that anyone does this, but it's interesting to read what's possible.
Here is a post by someone who created a Bitcoin seed generator powered by a smoke detector. They extracted Americium-241, hooked it to a Geiger counter and a Raspberry Pi to produce true random numbers.

There is a short video of this here:
https://x.com/pete_rizzo_/status/2087063715631837425

My 32 steel punch method is random.

8 characters is 32x32x32x32x32x32x32x32=1.099trillion set of combos. Or 40 bits [sound familiar] a single washer

16 characters is 1.099 trillion squared or 80 bits of entropy. A set of 2 washers.

24 characters is 1.099 trillion cubed or 120 bits of entropy. A set of 3 washers

32 characters is 1.099 trillion to the fourth 160 bits  of entropy. A set of 4 washers

40 characters is 1.099 trillion to the fifth 200 bits of entropy . A set of 5 washers

48 characters is 1.099 trillion to the sixth 240 bits of entropy. A set of 6 washers.

Simple punch
1
2
3
4
5
6
 On one side of the washers.

Then fill the other side with 8 characters.

[Danish Ali]

I'm saying that there may be something quite a bit deeper going on than a couple of thousand BTC being purloined.  At the very least the devices might be useful as evidence and would help paint a picture of their construction and distribution.

If the currently consensus story about a pre-processor macro being mis-used (deliberately or otherwise) is actually true, then the hardware itself may be worth something someday.  Making fairly safe original use of it would involve determining the purity of the hardware component supply chain, and compilation of the firmware by trusted parties (including oneself.)

If there are efforts to police up devices at the distributor or consumer level, that would be interesting to know of.

Never return the hardware when a potential back-door or flaw is involved.

A active attempt to recall or buy back these devices from retail channels would pretty much confirm that there's more to the story than just an 'unlucky macro bug'.

[philipma1957]

Fun fact: it takes less than 16.000 people that create a fresh wallet on a ColdCard device, to reach more than a 51% chance to stumble upon an already existing seed created by someone else. So, question of the day is: how many new wallets were created, over how many users, over how many sold devices, over 5+ years?

And this factors in boot time, menu interactions, and any USB activity. So imagine this: someone simply buys an affected device, creates his wallet, and boom, he's already rich. What would his next step be? Is he now a criminal? And who owns the BTC (the BTC, not the KYC sourced traceback funds)?

I know that in a group  of 30 people the odds are at 50 percent there are matching  birthdays.

So that is 30/365 for a match is: 50% +

In this case 16000 / 1.099 trillion for a match is: 51%+

By the way bing copilet says the odds of 16 ,000 picks from a set of 1.099 trillion  numbers having a collision is 1 in 8600

[vapourminer]

Fun fact: it takes less than 16.000 people that create a fresh wallet on a ColdCard device, to reach more than a 51% chance to stumble upon an already existing seed created by someone else. So, question of the day is: how many new wallets were created, over how many users, over how many sold devices, over 5+ years?

And this factors in boot time, menu interactions, and any USB activity. So imagine this: someone simply buys an affected device, creates his wallet, and boom, he's already rich. What would his next step be? Is he now a criminal? And who owns the BTC (the BTC, not the KYC sourced traceback funds)?

I know that in a group  of 30 people the odds are at 50 percent there are matching  birthdays.

So that is 30/365 for a match is: 50% +

In this case 16000 / 1.099 trillion for a match is: 51%+

birthday paradox

[philipma1957]

Fun fact: it takes less than 16.000 people that create a fresh wallet on a ColdCard device, to reach more than a 51% chance to stumble upon an already existing seed created by someone else. So, question of the day is: how many new wallets were created, over how many users, over how many sold devices, over 5+ years?

And this factors in boot time, menu interactions, and any USB activity. So imagine this: someone simply buys an affected device, creates his wallet, and boom, he's already rich. What would his next step be? Is he now a criminal? And who owns the BTC (the BTC, not the KYC sourced traceback funds)?

I know that in a group  of 30 people the odds are at 50 percent there are matching  birthdays.

So that is 30/365 for a match is: 50% +







In this case 16000 / 1.099 trillion for a match is: 51%+

birthday paradox

But I think he is wrong about 16000 picks of 1.099 trillion  is 50% for a collison.

16000 is 1 in 8600 bing
160000 is 1.16% google
1,600,000 is 65% google.

 My guess is they could have sold 160,000 wallets which would be under 2% of freak match

[JayJuanGee]

BTW, they are giving huge discounts now on all their Coldcard devices, does anyone want to buy it?

I wouldn't call 10% a huge discount. Whoever still buys that thing can't be helped.

I dunno.  The hardware is still, in my opinion and with limited analysis, the 'best in class'.  In my (somewhat limited) view, they had a lot of terrific ideas about security, and a knack for making the UI quite intuitive.  The documentation is fantastic, and the source code is available.  I just wouldn't be super confident that the binary actually represents the totality of the source code and it could be the case that they have some 'special patches'...I doubt that very many people availed of the feature of being able to build it oneself vs. just scoring one of Coinkit's release binaries.

Actually, I had some suspicions based on the observation that the product was almost 'too good.'  At least based on the common understanding of the size of their development team.

After a thorough analysis of their hardware, some care in sourcing, and a well structured and well know open source development team, the kit would be pretty compelling to me.  At least the Q.

ColdCard just caused the most catastrophic loss in the entire history of Bitcoin hardware wallets due to shockingly shoddy code, and the firmware fix for that caused some people's devices to brick, and the CEO spent the last 2 weeks deleting old posts on twitter as the company basically went into radio silence.... but other than that, Mrs. Lincoln, how was the play?

There's nothing special about ColdCard's hardware. It's cheap plastic. And some aspects of the design weren't even thought through, like the fact that the camera for the QR reader faces up, so you can't even see the screen while you try to scan a QR code. What the whut?

If you already own one and you want to accept the risks that come with staying with Coinkite... do what you think is best. Self custody comes with self-responsibility, so be honest with yourself about the risks. I have trouble imagining that company will even exist in a year, and there's no way I'd ever trust them. I wouldn't trust their hardware. I wouldn't trust their code. I wouldn't trust anything they say.

The level of their error in regards to the random number generator seems so outrageously basic that no one should be giving them any praise or talking positively about anything that they do, did or that they might do. There should be scepticism of their competency and their honesty.

I do understand that if the mistake had been made unintentionally and maybe even discovered after a bit of passage of time, then there may well would have had been a dilemma about what to do or how to handle it so that it was not taken advantage of by someone who would have heard the news and then exploited the guys who had already generated their keys through such greatly inferior random generator...

but yeah, holy fucking shit!!! The bug was introduced in March 2021 and yeah perhaps sloppily, and then there started to be reports of lost funds, which should have had caused the discovery of the bug within months of those reports of losses, especially if the ColdCard team was supposedly so competent and supposedly producing the most secure bitcoin wallet known to mankind.

[tvbcof]

The level of their error in regards to the random number generator seems so outrageously basic that no one should be giving them any praise or talking positively about anything that they do, did or that they might do. There should be scepticism of their competency and their honesty.

I do understand that if the mistake had been made unintentionally and maybe even discovered after a bit of passage of time, then there may well would have had been a dilemma about what to do or how to handle it so that it was not taken advantage of by someone who would have heard the news and then exploited the guys who had already generated their keys through such greatly inferior random generator...

but yeah, holy fucking shit!!! The bug was introduced in March 2021 and yeah perhaps sloppily, and then there started to be reports of lost funds, which should have had caused the discovery of the bug within months of those reports of losses, especially if the ColdCard team was supposedly so competent and supposedly producing the most secure bitcoin wallet known to mankind.

I respect your rigor and thoughtfulness on a lot of things, but with all due respect, I think you are wrong generally with respect to 'praise or talking positively' about a (potential) enemy.  It's a mistake which leads to losses a-la Sun Tzu and his 'know your enemy' for one thing.

Coldcard was seen as the 'luxury car' of cold storage devices among a lot of people including a lot of fairly technical people.  For a reason!

It is mighty suspicious that a (purportedly) small handful of people made a pretty wide variety of pretty good and pretty interesting gear, and yet 'failed' at the one very basic and very obvious thing (that being entropy of the private keys.)

If you blind yourself to the bigger picture out of spite/passion, you will very possibly miss some stuff about the big picture which is important.

[taufik123]

-snip-
but yeah, holy fucking shit!!! The bug was introduced in March 2021 and yeah perhaps sloppily, and then there started to be reports of lost funds, which should have had caused the discovery of the bug within months of those reports of losses, especially if the ColdCard team was supposedly so competent and supposedly producing the most secure bitcoin wallet known to mankind.

The actual bug could have been fixed long before this terrible hack happened, but those who gave the audit and notified a bug were instead ignored.
A Telegram user who first provided information about the bug in 2021, But it was simply ignored by Coinkite.
https://x.com/PanHodl/status/2084117965717323961/photo/1

And a second chance when X user James O'Beirne in May 2025 gave a warning to the problematic RNG Code, but again it was ignored. 
https://x.com/jamesob/status/2084624605969350915

And ignoring that important may not be without intent, as other users have found out about their own CTO's involvement,
with new evidence suggesting that the pseudonymous account that wrote the LibNgU code at the center of COLDCARD's entropy failure was actually Peter Gray, the founder and CTO of Coinkite.

[JayJuanGee]

The level of their error in regards to the random number generator seems so outrageously basic that no one should be giving them any praise or talking positively about anything that they do, did or that they might do. There should be scepticism of their competency and their honesty.

I do understand that if the mistake had been made unintentionally and maybe even discovered after a bit of passage of time, then there may well would have had been a dilemma about what to do or how to handle it so that it was not taken advantage of by someone who would have heard the news and then exploited the guys who had already generated their keys through such greatly inferior random generator...

but yeah, holy fucking shit!!! The bug was introduced in March 2021 and yeah perhaps sloppily, and then there started to be reports of lost funds, which should have had caused the discovery of the bug within months of those reports of losses, especially if the ColdCard team was supposedly so competent and supposedly producing the most secure bitcoin wallet known to mankind.

I respect your rigor and thoughtfulness on a lot of things, but with all due respect, I think you are wrong generally with respect to 'praise or talking positively' about a (potential) enemy.  It's a mistake which leads to losses a-la Sun Tzu and his 'know your enemy' for one thing.

I am not really talking about them in terms of being an enemy, and so each of us have rights to feel various kinds of attachment, and from my perspective, in this particular case, and based on your last few posts, you seem to have Stockholm syndrome.

Of course, so far, I have not heard of any criminal charges being brought against Cold Card or any of their principles, and I have not heard of any significant lawsuits, even though I would think some lawsuits would be coming, since there seems to be quite a bit of evidence showing either malice or such a high level of recklessness that they may well end up having to answer for in more formal ways, rather than just responses on the internet and trying to salvage whatever amount of good will that might still exist from those who are not realizing the outrageousness (and unforgivable) place for such a high level of mistake.

I do understand that in bitcoin there are all kinds of attacks on it, and so we cannot choose who participates in bitcoin and/or how they choose to participate in bitcoin,  yet on personal levels we can express our concerns about some players and even our concerns about some members who might be supporting malicious and/or bad intended actors in the space.

We can disagree too, so even though I think that it is problematic to be saying anything nice about Cold Card, you have a right to say it, and you are likely to get push back from various members on the forum, including yours truly.

Coldcard was seen as the 'luxury car' of cold storage devices among a lot of people including a lot of fairly technical people.  For a reason!

I suppose I am not against recognizing some of their technical innovations as long as you are sure to call them twats at the same time... and yeah, I am not trying to tell you what to say, since you can do what you like. 

It might be my own current perspective, yet if we want to compare current Cold Card advocates to shitcoiners, then that may well be a fair framework since there are some shitcoins that might have some bells and whistles that might bring some innovation to the space, perhaps? 

It is mighty suspicious that a (purportedly) small handful of people made a pretty wide variety of pretty good and pretty interesting gear, and yet 'failed' at the one very basic and very obvious thing (that being entropy of the private keys.)

You are not going to get argument from me about that characterization based on what seems to be currently known.

If you blind yourself to the bigger picture out of spite, you will very possibly miss some stuff about the big picture which is important.

I don't think that I am being blind based on my comment that they do not deserve any praise that is not coupled with with calling them names at the same time.  Sure, I might be exaggerating a bit, since I am not going to necessarily going to back off on the contents of my earlier post. 

It was pretty outrageous what they did and sure it is possible that only one or two of them are at fault, perhaps?  Yet at the same time, I think that the whole situation is still under investigation, to the extent that evidence has not been destroyed, even past communications being deleted which is a pretty cowardly (and even guilt-demonstrating) kind of behavior, from my perspective, even if they might be emotional at the moment, they likely have obligations to the bitcoin space based on how many bitcoin hardware wallets and other products they ended up selling and also the fact that they had a lot of prominent bitcoiners singing their tunes, whether those prominent bitcoiners were paid or not, there was a level of contagion in the marketing that they likely did not deserve, so it seems problematic if guys are still repeating some of those supposedly positive attributes, even if you are claiming them as supposed facts.

I doubt that it is productive for us to go one by over some supposed Cold Card innovation, to the extent that it is relevant right now when many guys are still just finding out about their losses and they are going to continue to come in, to the extent that new waves of attacks on poorly secured wallets (that had been recently broken) might also come in, too.

By the way, at any point in time, any one of us might have certain opinions and/or feelings (including myself), and I might change my mind later down the road, and I don't expect to have all of the information, and surely I don't expect to even know much about various technical prowesses that Cold Card might have had, yet at the same time, I can choose my priorities in terms of what to emphasize, which right now continues to include a certain level of outrage about their conduct and even sadness in the ongoing reactions and even really smart people describing their discoveries of what had been going on with Cold Card and becoming emotional in their attempts to objectively describe the events as they are unravelling and how we are continuing to hear about different angles of the inside scoop and even hearing about both losses and other attempts of the community incluing the red team that had spawn out of this.

[tvbcof]

I am not really talking about them in terms of being an enemy, and so each of us have rights to feel various kinds of attachment, and from my perspective, in this particular case, and based on your last few posts, you seem to have Stockholm syndrome.

I am certainly thinking and talking of Coinkite as the enemy for reasons I won't go into at the moment.  In fact, a bitter enemy who I would like to see suffer greatly.  Right now I don't know who 'them' is though, and it is rare for certain classes of criminals to become accused, indicted, prosecuted, or covicted so I don't wait for these to occur before considering them an enemy.

I have to say that you are probably mis-reading, or not reading deeply enough into my writings as I intended if you think I have 'Stackholm syndrome' (or any such psychological condition which would cause me to 'defend' them.)  That is understandable in part because I have an older style of writing, and I also like to try to look at and describe things from all angles as a method of analyzing things.

...
We can disagree too, so even though I think that it is problematic to be saying anything nice about Cold Card, you have a right to say it, and you are likely to get push back from various members on the forum, including yours truly.

To me, pushback is a good thing as it gets everyone thinking more.

Thanks for the rest of your reply.

[bitbollo]

After this disaster, I would not suggest to anyone to use any of their products (with all due respect to people involved).
I would get their hardware as COLLECTIBLE (likewise I have a collection of these items from many producers - mostly SEALED)... anyway these funds/products must be used to compensate people damaged or the offer a serious research on this issue.
Please note that any seed generated from this hardware from March 2021 to now could be considered COMPROMISED.

...
Coldcard was seen as the 'luxury car' of cold storage devices among a lot of people including a lot of fairly technical people.  For a reason!
...

It could be seen as pretty questionable that this product could be seen as "luxury car" of cold storage.
Many users involved with bitcoin have confirmed the same point= would you trust your bitcoin with a seed/private key produced by a third part?
Well, I can't stand this group of "well knowledge fairly technical people" but it's evident that you don't need both coldcard or an hardware wallet to store your coins or to use bitcoin.

RE: How many possibly bitcoin addresses are there exactly? And how long does it...
https://bitcointalk.org/index.php?topic=24268.msg41719898#msg41719898