Ex–Mt. Gox CEO Mark Karpelès Proposes Bitcoin Hard Fork to Reclaim 80,000 Stolen BTC

Crypto Twitter (CT) loves a good resurrection arc. This week's candidate is Mark Karpelès, the former CEO of Mt. Gox, walking back into the Bitcoin$62,492.80 discourse with a spicy ask: change Bitcoin$62,492.80's rules so a known stash of stolen coins can be reclaimed. ^[1]

The key fact: Karpelès submitted a proposal on GitHub that would add a new Bitcoin$62,492.80 consensus rule to redirect 79,956 Bitcoin linked to the Mt. Gox hack, coins that have sat for years in a single address (commonly tracked as 1FeexV6bAHb8ybZjqQMjJrcCrHGW9sb6uF). At roughly current prices, that is about $5.2 billion worth of Bitcoin. His framing is simple, it has been more than a decade, the bankruptcy has dragged on, and this is one of the last unresolved pain points.

Citi Targets 2026 Launch of $30T Custody Rails as Bitcoin Integration Accelerates

If you hear "hard fork" and your first thought is "surely Bitcoin will never," you are not alone.

Enjoy articles without ads?

What Karpelès actually proposed

Karpelès' GitHub pull request proposes a rule change that would allow a specific set of coins, identified by their history and current unspent outputs, to be spent to a designated "recovery address" without the original private key.

That last part is the entire controversy. Bitcoin ownership is enforced by cryptography plus consensus: if you can produce a valid signature, the network accepts the spend. If you cannot, you do not get to move the coins, even if everyone "knows" they are stolen.

Karpelès' idea effectively introduces a one-off exception at the protocol level. Put differently: the chain would start treating a particular set of coins as encumbered by a new rule that overrides the normal signature requirement, but only for the purpose of transferring them to a recovery destination.

Hard fork, defined (and why this one is hard)

A hard fork is a change to the protocol rules that is not backward compatible. Nodes that do not upgrade will reject blocks produced under the new rules. In this case, allowing coins to move without a valid signature would make some transactions valid that were previously invalid, which is classic hard fork territory.

Bitcoin has historically avoided contentious hard forks because they risk splitting the network into two competing versions of "Bitcoin," plus all the chaos that comes with it: exchange listings, replay protection, miner alignment, and a long tail of user confusion.

Starknet Set to Launch 'Private Bitcoin' Layer With Zcash-Style Shielded BTC Transfers

Why CT is already side eyeing it

The proposal lands on a cultural fault line that Bitcoiners treat as almost sacred: immutability (the idea that the ledger's history and rules are not up for ad hoc edits) and fungibility (one Bitcoin should be as "clean" as any other).

A few reasons the pushback is predictable:

1) "Code is law" runs both ways

If Bitcoin can be modified to recover stolen coins here, it signals that the rules can be modified again for other cases later. Even people who sympathize with Mt. Gox victims worry about the precedent: today it is hacked exchange coins, tomorrow it is sanctions, court orders, political pressure, or simple majoritarian vibes.

Bitcoin's social contract is already complicated. Adding a "special case" exception is gasoline on a very old fire.

2) Who decides what is stolen?

The proposal hinges on strong attribution: that the coins in that specific address are indeed hack proceeds and should be forcibly redirected.

But "everyone knows" is not a cryptographic primitive. For Bitcoin to bless that attribution at the protocol layer, the network would be turning a social and legal judgment into a consensus rule. Bitcoin's maintainers and node operators have historically resisted becoming arbiters of property disputes for exactly this reason.

3) Fungibility takes a hit

If the network can tag certain UTXOs (unspent transaction outputs) as recoverable, it encourages the idea of "tainted" coins with different rules. That is a fast path to more surveillance, more blacklists, and a less neutral base layer, even if the initial target is universally unpopular thieves.

4) It likely goes nowhere without overwhelming consensus

Bitcoin development is conservative by design. A GitHub pull request is not "a fork," it is a starting point for review, debate, and usually rejection unless there is broad agreement across developers, miners, businesses, and node operators.

The vibe check here is brutal: absent near unanimous support, a hard fork is basically dead on arrival.

The Mt. Gox context that still haunts the timeline

Mt. Gox was not a small exchange that imploded quietly. At its peak, it handled a huge share of global Bitcoin trading. When it collapsed in 2014, reports centered on the loss of roughly 850,000 Bitcoin, later offset by about 200,000 Bitcoin recovered, leaving a massive hole that shaped early Bitcoin's collective trauma. ^[2]

Karpelès has remained a controversial figure ever since. Japanese courts ultimately cleared him of some of the most serious allegations (notably embezzlement) while convicting him on data related charges, a legal outcome that did not exactly produce closure for creditors or the broader community. ^[3]

Meanwhile, the bankruptcy and rehabilitation process has crawled forward for years, with creditors tracking wallets and court updates like it is a long running ARG. Against that backdrop, the idea of "just fork it back" will inevitably attract attention, even if only as discourse fuel. ^[4]

The deeper issue: Bitcoin is not Ethereum, and 2016 is not 2026

People will inevitably compare this to Ethereum$1,686.33's 2016 DAO rollback, where the community opted to fork to reverse a massive exploit, creating today's Ethereum$1,686.33 and Ethereum Classic$8.64 split.

But Bitcoin's culture and governance are different. Ethereum$1,686.33's community has historically been more willing to coordinate social recovery in exceptional cases. Bitcoin's base layer identity is tied to resisting exactly that kind of discretionary change. That does not make one side morally superior, but it does make the probability of a Bitcoin "recovery fork" extremely low.

Also, the ask here is narrower and arguably more surgical than a full rollback, but it is still a protocol level override of signature based ownership. To many Bitcoiners, that is the line.

BTC and ETH Face $8.72B Options Expiry: Volatility Spikes as Traders Watch for a Max-Pain 'Pain Trade'

What to watch next (and what risks matter)

If you are holding a bag or just watching the spectacle, the practical checklist is straightforward:

GitHub and developer discussion: Watch whether the proposal gains any serious technical review or is quickly dismissed as out of scope. Bitcoin Core contributions can spark debate even when they never merge.
Node and miner sentiment: A hard fork without broad operator buy in is not "an upgrade," it is a chain split risk.
Legal and creditor developments: Mt. Gox repayments and court processes are still the legitimate pathway for most creditors. A protocol intervention would not automatically map cleanly onto legal ownership anyway.
On chain movement of the 1Feex address: Any movement from the long watched wallet would be a bigger market signal than another round of GitHub comments.

Bottom line: Karpelès put a meme sized idea into a very serious place, Bitcoin consensus rules. The odds of Bitcoin rewriting its signature guarantees to rescue a specific pool of stolen coins are slim, but the conversation is revealing. It is a reminder that "immutability" is not just a technical property, it is a social choice that Bitcoiners re affirm every time they refuse to make an exception.