The Ethereum layer-2 network Blast is shutting down, leaving roughly $51 million in user funds on the platform and an unresolved question of who controls those assets as the network winds down, BeInCrypto reports.

The pending shutdown puts a spotlight on the custody architecture of a network that grew rapidly after its public launch. Blast, developed by the team behind the non-fungible token marketplace Blur, attracted deposits by offering yield on bridged ETH and stablecoins before its mainnet went live. That design funneled user assets into the network's bridges, where returns accrued from staking and token incentives.

With the network now closing, the remaining $51 million sits in contracts whose control depends on the project's multisig signers and bridge infrastructure. Users who fail to withdraw before final shutdown steps risk depending on those custodial arrangements to recover funds.

The wind-down illustrates a broader operational risk in the layer-2 sector: networks that operate through upgradeable contracts and multisig-controlled bridges can leave depositors exposed when teams disband. Bridge withdrawals typically remain possible only while sequencers, validators or the underlying infrastructure stay online, and users of defunct rollups have historically needed to execute trust-minimized exits before a network's contracts are frozen or deprecated.

Blast's trajectory was unusually fast. The network reported billions of dollars in total value locked within weeks of opening deposits in late 2023, driven by its points-based airdrop campaign, before activity and inflows cooled. The decision to shut down follows that decline in usage.

For remaining depositors, the operational priority is straightforward: verify withdrawal routes, move assets to the Ethereum mainnet or another supported destination, and avoid leaving balances in bridge contracts that may lose support once the team stops maintaining infrastructure. As the shutdown proceeds, the fate of the final $51 million will test whether Blast's exit process treats user custody as a first-order obligation or an afterthought.