Proposal: Consensus rule to recover 79,956 BTC stolen from MtGox

[MagicalTux]

I've opened a pull request on Bitcoin Core to discuss a one-time consensus rule that would allow the ~79,956 BTC stolen in the 2011 MtGox hack to be recovered and returned to creditors.

Pull request: bitcoin/bitcoin#34695

Context

For those who weren't around or need a refresher: in June 2011, a compromise of MtGox systems resulted in ~79,956 BTC being sent to 1FeexV6bAHb8ybZjqQMjJrcCrHGW9sb6uF. These coins have not moved in over 15 years. The UTXO is still sitting there, along with thousands of dust outputs people have sent to it over the years.

MtGox is currently in civil rehabilitation under Japanese courts. Trustee Nobuaki Kobayashi has been distributing recovered funds to creditors, but these 79,956 BTC are out of reach -- nobody on the creditor side has the private key, and the attacker has shown no sign of moving them in a decade and a half.

What the proposal does

The patch adds a new script verification flag that, when active, allows outputs paying to the theft address to be spent using a signature from a designated recovery address. It is gated behind an activation height that is currently set to INT_MAX (i.e. it does nothing until the community agrees on a real activation height).

The change is ~50 lines of code. It affects one specific P2PKH address and no other consensus rules.

Why I think this is worth discussing

I know the first reaction for many will be "this violates immutability" or "where does it end." Those are legitimate concerns, and I'm not dismissing them. But I think this case has a combination of factors that make it genuinely unique:

• Unambiguous theft. This isn't a dispute about ownership. Law enforcement in multiple jurisdictions confirmed the hack. There is no question these coins were stolen.
• 15+ years dormant. The coins haven't moved since 2011. The keys may well be lost entirely, making these coins effectively burned. Recovery would return dead coins to productive economic use.
• Existing distribution framework. A Japanese court-supervised rehabilitation process is already actively paying creditors. The legal infrastructure to handle these coins is in place and has been running for years. This isn't a situation where we'd be recovering coins with no clear destination.
• Narrowly scoped. This targets exactly one address. It cannot be extended to other cases without a separate, independent consensus change that would need its own community agreement.
• UTXO cleanup. As a side benefit, the address has accumulated a lot of dust. Sweeping it cleans up the UTXO set for every node operator.

Why now

The MtGox trustee has declined to pursue on-chain recovery because there's no certainty it would be adopted. But the community can't evaluate the idea without a concrete proposal. This creates a deadlock. I'm breaking that deadlock by putting actual code on the table so there's something concrete to discuss rather than hypotheticals.

I'm under no illusion that this would be merged quickly or without extensive debate. That debate is the entire point.

The strongest argument against

I want to be honest about this: the strongest argument against is precedent. If Bitcoin makes an exception once, it becomes harder to refuse the next request, and the next. The line between "unique, justified exception" and "political override of property rules" is subjective and hard to maintain.

I believe this specific case is different enough -- in its documentation, its age, its legal framework, and its scope -- that it can stand on its own without opening floodgates. But I understand if others disagree, and I respect that position.

This is a hard fork

I want to be upfront: this is a hard fork. It makes a previously invalid transaction valid. All nodes would need to upgrade before the activation height. I'm not trying to disguise that fact or sneak it through as something else.

Open questions

Even if the community is open to the idea, there are design questions:

• What activation mechanism? Fixed height, BIP9 miner signaling, or something else?
• Should there be a time-limited recovery window (e.g. 1 year) after which the rule expires?

I'm here to discuss all of this. Fire away.

-- Mark Karpeles

[stompix]

You've got to be kidding me!

[MagicalTux]

You've got to be kidding me!

Just trying to open a discussion here. We've seen something similar with ethereum before (with the DAO thing) which caused a fork, but in the end consensus regrouped around the "right thing" which is letting people who were hurt be saved.

I know this is not going to be an easy proposal, but I do believe there is strong arguments toward doing it. There are arguments against too, so I'm hoping to see what everyone thinks and weigh the pro/cons.

[Cookdata]

You've got to be kidding me!

Just trying to open a discussion here. We've seen something similar with ethereum before (with the DAO thing) which caused a fork, but in the end consensus regrouped around the "right thing" which is letting people who were hurt be saved.

I know this is not going to be an easy proposal, but I do believe there is strong arguments toward doing it. There are arguments against too, so I'm hoping to see what everyone thinks and weigh the pro/cons.

Bitcoin isn't ethereum, no way this is getting accepted. No way!

[paraboul]

You've got to be kidding me!

Just trying to open a discussion here. We've seen something similar with ethereum before (with the DAO thing) which caused a fork, but in the end consensus regrouped around the "right thing" which is letting people who were hurt be saved.

I know this is not going to be an easy proposal, but I do believe there is strong arguments toward doing it. There are arguments against too, so I'm hoping to see what everyone thinks and weigh the pro/cons.

Bitcoin isn't ethereum, no way this is getting accepted. No way!

Ok, but what's the argument here, besides 'it's not Ethereum'?

[coupable]

You've got to be kidding me!

Just trying to open a discussion here. We've seen something similar with ethereum before (with the DAO thing) which caused a fork, but in the end consensus regrouped around the "right thing" which is letting people who were hurt be saved.

I know this is not going to be an easy proposal, but I do believe there is strong arguments toward doing it. There are arguments against too, so I'm hoping to see what everyone thinks and weigh the pro/cons.

Bitcoin isn't ethereum, no way this is getting accepted. No way!

Some others will suggest that dormant addresses belong to Satoshi can be added to the pool of recovered bitcoin. Then later, each time a hack incident will happe, someone will call for another new consensus rule to recover stolen funds. This will destroy the bitcoin concept in full. No way this will be accepted except for MYgox victims who are few people considering that time when the incident happen.

[MagicalTux]

Bitcoin isn't ethereum, no way this is getting accepted. No way!

What would be your strongest argument against? Those bitcoins have been sitting these for 15 years now, everyone knows these are from the MtGox hack, yet some people have tried to use underhanded methods to claim these using courts/etc. Allowing these to be returned to the MtGox customers would make a lot more sense, have a minimum cost and, I believe, would be the right thing to do.

[Knight Hider]

No. Create your own fork if you want this.

[MagicalTux]

Some others will suggest that dormant addresses belong to Satoshi can be added to the pool of recovered bitcoin. Then later, each time a hack incident will happe, someone will call for another new consensus rule to recover stolen funds. This will destroy the bitcoin concept in full. No way this will be accepted except for MYgox victims who are few people considering that time when the incident happen.

This case is quite unique for multiple reasons.

The bitcoins haven't moved, can be directly linked to the theft (those do not belong to Satoshi but to the mtgox hack itself). The coins have been tagged, and the hacker clearly knows moving these would trigger news worldwide instantly.

There aren't many events that affected so many Bitcoin users as MtGox (arguably most bitcoin users had an account on MTGox in 2014), where we both know which bitcoins are the stolen bitcoins and have a legal process that would allow distribution of said coins back to creditors.

[stwenhao]

This is a hard fork

Then just do that. You will have just yet another altcoin. Why do you need to get Core's approval? All existing nodes will reject your hard-fork anyway, even if Core would release a new version with your changes.

What activation mechanism?

None. Activation mechanisms are for soft-forks. For hard-forks, you just release a new version, and you are good to go. Welcome to the altcoin world!

Should there be a time-limited recovery window (e.g. 1 year) after which the rule expires?

It will be rejected by all existing nodes, so nobody cares. You can do whatever you want. It doesn't matter if you allow moving coins without valid signatures, or if you double the supply. Hard-fork will kick you to the altcoin world anyway.

You've got to be kidding me!

Seems like some people are celebrating 1st of April one month earlier.

[1440000bytes]

This is a hard fork

I want to be upfront: this is a hard fork. It makes a previously invalid transaction valid. All nodes would need to upgrade before the activation height. I'm not trying to disguise that fact or sneak it through as something else.

Open questions

Even if the community is open to the idea, there are design questions:

• What activation mechanism? Fixed height, BIP9 miner signaling, or something else?
• Should there be a time-limited recovery window (e.g. 1 year) after which the rule expires?

I'm here to discuss all of this. Fire away.

-- Mark Karpeles

- Try BIP 9 miner signaling and see how many mining pools would run the activation client. If enough pools participate, it could become another airdrop for everyone not just Mt. Gox customers.

- The recovery window should be unlimited.

[goldkingcoiner]

The day this becomes possible, I am out.

Good thing that it is never going to happen. No matter how moral of a reason, Bitcoin is not and should never be something that can get taken away out of a wallet or recycled because it has been considered "lost", "stolen" or "dust". This will give leeway to misuse and corruption.

That kind of thing goes too far.

Not your keys, not your coins.

[paraboul]

This is a hard fork

Then just do that. You will have just yet another altcoin. Why do you need to get Core's approval? All existing nodes will reject your hard-fork anyway, even if Core would release a new version with your changes.

What activation mechanism?

None. Activation mechanisms are for soft-forks. For hard-forks, you just release a new version, and you are good to go. Welcome to the altcoin world!

Should there be a time-limited recovery window (e.g. 1 year) after which the rule expires?

It will be rejected by all existing nodes, so nobody cares. You can do whatever you want. It doesn't matter if you allow moving coins without valid signatures, or if you double the supply. Hard-fork will kick you to the altcoin world anyway.

You're underestimating how many people have skin in the game here  

[stompix]

There are arguments against too, so I'm hoping to see what everyone thinks and weigh the pro/cons.

Oh no shit!
Let's first assume you are indeed the one you claim to be!
We have a liar, a fraud, a convict, a man who managed to destroy the reputation of the Bitcoin Foundation and then bring ruin to thousands with his incompetence, coming with a proposal that would destroy the very foundation of Bitcoin!

I repeat myself, are you fucking kidding me right now?

Moving forward with this, don't even bother with anything else because this is the most important thing
- What would happen with the recovered BTC and how much of it you want to stuff in your pockets?

[Z-tight]

What would be your strongest argument against

You have already covered all the strong arguments against it in your op. To put it in the simplest way possible, it goes against what BTC is and this will be rejected by all full node operators. It is not even recovery, it is theft.

More exchanges than anyone can count have been hacked and lost customer funds since the MtGox incident, those are also 'unambiguous theft', if i can take that line from your op, so why not recover all of that too. What of retail investors losses?
Whoever controls the keys in the BTC network spends the coins and that's that.

[MagicalTux]

Oh no shit!
Let's first assume you are indeed the one you claim to be!
We have a liar, a fraud, a convict, a man who managed to destroy the reputation of the Bitcoin Foundation and then bring ruin to thousands with his incompetence, coming with a proposal that would destroy the very foundation of Bitcoin!

I repeat myself, are you fucking kidding me right now?

I beg to differ. Yes I have been found guilty of hiding the massively lacking bitcoins. That is under Japan's legal regime where 99% of charges pressed lead to a guilty verdict. The Bitcoin Foundation didn't need help to go down (and I wasn't involved there anyway).

I deeply regret what went on and trying to do everything I can to make creditors whole. This is one of the things I can do. Will it yield to something? Only way to know is to do it. Hate me all you want.

Moving forward with this, don't even bother with anything else because this is the most important thing
- What would happen with the recovered BTC and how much of it you want to stuff in your pockets?

Recovered BTC would be under supervision of the court and the trustee until distributed to creditors. I would get exactly wero of these.

[MagicalTux]

What would be your strongest argument against

You have already covered all the strong arguments against it in your op. To put it in the simplest way possible, it goes against what BTC is and this will be rejected by all full node operators. It is not even recovery, it is theft.

More exchanges than anyone can count have been hacked and lost customer funds since the MtGox incident, those are also 'unambiguous theft', if i can take that line from your op, so why not recover all of that too. What of retail investors losses?
Whoever controls the keys in the BTC network spends the coins and that's that.

Can you point to any unambiguous theft proceeds still on chain on a single easily recovered address?

[Cookdata]

Bitcoin isn't ethereum, no way this is getting accepted. No way!

What would be your strongest argument against? Those bitcoins have been sitting these for 15 years now, everyone knows these are from the MtGox hack, yet some people have tried to use underhanded methods to claim these using courts/etc. Allowing these to be returned to the MtGox customers would make a lot more sense, have a minimum cost and, I believe, would be the right thing to do.

And what would you have to do about other Bitcoin hacks that happened after then? I'm not sure if Mt gox was the first exchange to be a victim of hack, last year alone Bybit was hacked roughly $1B+ and nothing was done, those are literally users funds, the exchange is still on debt as they have to borrow Bitcoin from other institutions to cover up the loss; this is just one I can remembered for now.


If I'm a victim of hack, I will be so pained too but Bitcoin doesn't know what is wrong and what's right, only human are judgmental about what's wrong and right which is subjective, you don't get to decide what's right and what's wrong when it comes to Bitcoin, if you own it, prove it.
With this proposal, only the community can decide on what happen next but deep down you know the outcome, it's going to be rejected.



This is you? Another food for knots community, they told you to run knots on the comment section. 

[coupable]

Some others will suggest that dormant addresses belong to Satoshi can be added to the pool of recovered bitcoin. Then later, each time a hack incident will happe, someone will call for another new consensus rule to recover stolen funds. This will destroy the bitcoin concept in full. No way this will be accepted except for MYgox victims who are few people considering that time when the incident happen.

This case is quite unique for multiple reasons.

The bitcoins haven't moved, can be directly linked to the theft (those do not belong to Satoshi but to the mtgox hack itself). The coins have been tagged, and the hacker clearly knows moving these would trigger news worldwide instantly.

How are you sure that that address really belongs to hacker? It's a great probability but still not 100% sure about this. Maybe time will tell. The hacker can use mixers or even create his own mixer to bury all logs and run away with clean bitcoins. Nothing can stop him from doing so if the address really belong to him, or maybe he is waiting for an exact date. All scenarios are possible.

There aren't many events that affected so many Bitcoin users as MtGox (arguably most bitcoin users had an account on MTGox in 2014), where we both know which bitcoins are the stolen bitcoins and have a legal process that would allow distribution of said coins back to creditors.

There might be events in the future that will affect so many bitcoin usersmore than MTgox incident. Should we then suggest a new consensus rule to recover them as well?

[stwenhao]

You're underestimating how many people have skin in the game here

Good luck. If someone will make a hard-fork, then people will just sell coins they don't want to, and buy the coins they want. If it is a hard-fork, then it is handled in the same way as every other altcoin. It is just repeated 2017 with BCH, and all of that.

Which means, that you don't need any activation, any signalling, nothing. You release a version, that would be hard-rejected by all existing infrastructure anyway, no matter who will release a new version. So, what do you want to signal by BIP-9 or anything else into existing nodes? The fact, that you are making a chain, which old nodes will never accept? All altcoins could send the same signal, and it wouldn't bring any value.