A Polygon governance proposal would hard-fork the chain to move 70.49M POL from five blocked attacker addresses to Polymarket and Binance. Its authors call it an extreme action.
The stolen POL can’t move, and under Polygon’s current setup it can’t go home either. A proposal on the Polygon governance forum asks validators to fix that by rewriting five balances on a live chain, then warns in the same breath that nobody should treat it as a template.
Five Frozen Addresses and 70.49M POL
The proposal went up on November 4 and is on the agenda for today’s Polygon Protocol Governance Call at 14:30 UTC. It follows separate security incidents at Polymarket and Binance on October 29 and 30. Stolen native POL landed on addresses on Polygon PoS, and at the victims’ request the block producers added those addresses to the Bor transaction block list.
Five attacker-controlled addresses now hold about 70.49M POL, roughly $6.5M at the PIP’s reference price of $0.0924. Four of them, holding about 23.86M POL, trace to Polymarket. The fifth, holding about 46.63M POL, traces to Binance. Polymarket and Binance are named as the proposers.
How the Funds Reached the Blocked Wallets
The proposal lays out the inflow path in enough detail to follow on a block explorer.
On the Polymarket side, the largest address, 0xe4C4…1EC6, held 15,331,308.29 POL at mainnet block 96,399,251. It also sent 3,282,275.71 POL onward to a second address in two native transfers at 18:30:34 and 18:35:59 UTC on October 29. Two other addresses hold about 4.3 million and 4.2 million POL, and the fourth now holds almost nothing.
On the Binance side, a Binance hot wallet sent 46,626,589.6 POL and 6,185,900.1 USDT to 0x5227…2f8c in block 96,406,484. Tether has since frozen the USDT, which leaves the native POL as the balance still at risk. Polymarket has published its own post-mortem.
That traceability strengthens the case. The argument for the fork rests on transfers anyone can inspect, not on anyone’s say-so.
Why the Block List Is Only a Stopgap
The proposal is unusually candid about the limits of the current defense. Each block producer reads its block list only at startup, so every change needs a deployment and a Bor restart or span rotation. Entries have no automatic expiry, and the filter covers only outer senders and EIP-7702 authorities, which means it isn’t a complete asset freeze.
Meanwhile, the attacker hasn’t gone quiet. A pending transaction with a fee of about 219,450 POL, or 10.45 POL per gas, was observed on October 30. The attacker has also contacted validators offering cash to help unblock a transaction. Polymarket opened an on-chain recovery channel to the attacker through the SEAL Safe Harbor Recovery safe and has received no response.
This detail carries the most weight. A defense that depends on every producer restarting with an unmodified file, against an attacker who is bribing validators, is not a stable end state.
What the Hard Fork Would Change at Block F
On paper, the fix is narrow. At the start of an activation block, before any transaction in it runs, the client would set each listed attacker address to zero and credit the same amount to a recovery address designated by Polymarket or Binance. Total supply is unchanged. Heimdall, the StakeManager and the bridge contracts aren’t touched, and only native POL balances in the listed addresses are affected.
The mechanics carry some practical consequences. Because the change isn’t a transaction, explorers, indexers and accounting tools that rebuild balances from transactions will need to handle it as a special case. Nodes that don’t upgrade will follow a different chain from that block. The proposal estimates about two weeks after approval as workable for client development and an Amoy testnet rehearsal, and it expects 70 to 80% of stake to reach the required version quickly after release. Each recovery address must also prove control before activation.
The Conditions That Make This an Exception
Rather than asking for trust, the authors lay out a checklist of conditions. The hacks are publicly acknowledged by the victims. Every block producer has blocked the addresses. The amount is meaningful. The funds trace on-chain to known victims, who are also the ones requesting the remedy. And no party with a possible claim has contested the account of events.
The proposal calls the move an “extreme action” and says it relies on all of those conditions holding together. A future request that doesn’t meet them, it says, shouldn’t cite this PIP as precedent. It also rejects a general freeze-and-recover mechanism, since that would give the protocol a permanent power over user funds. For now, Binance’s smart-contract hot wallet and other additional addresses are left out of the sweep.
The Precedent Debate Heading Into Governance
The early forum replies show the split. One supporter argues that even decentralized systems need collective action in exceptional circumstances. Another participant asks for a published activation block and a machine-readable address list so exchanges and custodians can reconcile their records.
The sharpest reply questions the precedent itself. It warns that once Polygon shows native balances can be reassigned, courts and law enforcement may ask for the same treatment, and it asks who would separate legitimate recovery from politically motivated seizure. That question sits at the heart of the debate. Ethereum went through it with its 2016 DAO fork, which returned funds but split the community, and the Polygon authors have tried to engineer around the same problem with narrow conditions.
In the end, the proposal will succeed or fail on whether validators believe those limits will hold. The details to watch are the final address list, the snapshot-block balances and the activation date. The first test comes today on the governance call.

