Polygon proposes hard fork to recover $7M in stolen Binance, Polymarket funds
Belle Femme Emmo / Wikimedia Commons (CC BY-SA 4.0)
A new governance proposal would zero out five attacker-controlled addresses and send roughly 70.49 million POL back to verified recovery accounts
Polygon developers want to do something blockchains rarely do: rewrite balances on purpose. A new Polygon Improvement Proposal (PIP) calls for a hard fork of the Polygon Proof-of-Stake (PoS) mainnet to claw back stolen tokens tied to separate security incidents at Binance and Polymarket.
The target is roughly 70.49 million native POL, worth around $7 million, sitting in five attacker-controlled addresses. The tokens are already frozen. The fork would move them back to their rightful owners.
What the proposal would actually do
The PIP was introduced on November 4, 2026, by Parvez03, a member of the Polygon Labs validator support team. It describes an exceptional state change to the network’s ledger, not a routine upgrade.
The stolen funds trace back to two incidents on October 29 and 30. About 23.86 million POL ended up across four addresses connected to the Polymarket breach. Another roughly 46.63 million POL sits in a single address linked to the Binance incident.
Under the proposal, all five addresses would have their balances set to zero. Matching amounts would then be credited to verified recovery accounts set up for each victim.
The total supply of POL would not change. No new tokens get minted, and none get burned. The coins simply change hands, from the thieves back to the people they were taken from.
Frozen, but not finished
Polygon’s block producers have already added the five addresses to a transaction denylist. That blocks the funds from moving, at least in theory.
The attackers have not exactly given up. According to the proposal’s backers, they keep trying to push the assets out through high-fee transactions and alleged bribery of validators.
The news moving money, markets, and the world—before your day starts.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
Developer teams have stressed that the measure is extraordinary. They say it is not meant to set a precedent for future incidents.
The scope is also narrow. The recovery covers only the native POL still locked on Polygon PoS. Other tokens involved in the incidents, including USDT, fall outside this proposal.
Polymarket and Binance have already reimbursed most affected users on their own. So the fork is less about making individual users whole and more about recovering the remaining stolen assets that are stuck on-chain.
The timeline from here
The community is set to discuss the PIP at the Protocol Governance Call on November 5, 2026, at 14:30 UTC.
If it gets the green light, client development and validation on the Amoy testnet are expected to take around two weeks before the change could go live on mainnet.
Why a reversal is a big deal
Immutability is one of the core selling points of public blockchains. Once a transaction is confirmed, it is supposed to stay confirmed. That promise is what lets strangers trust a ledger nobody controls.
Polygon’s framing tries to manage that tension. By routing the change through its formal governance process, labeling it exceptional, and keeping total supply untouched, the team is signaling that this is a carefully bounded intervention rather than a new habit.
What this means for POL holders and the wider market
The supply-neutral design removes one obvious market concern. Because no POL is created or destroyed, the fork itself should not dilute existing holders. The tokens returning to victims were already part of the circulating picture.
For Binance and Polymarket, recovery would offset funds they have largely already covered out of pocket for users. That makes the fork a balance-sheet matter as much as a user-protection one. Recovering the assets on-chain could ease the cost of those reimbursements.
Disclosure: This article was edited by John Chen. For more information on how we create and review content, see our Editorial Policy.

