Moving an asset between blockchains has usually meant trusting someone in the middle. A bridge holds the deposited funds, or a custodian holds the keys, and either becomes a single point of failure and a target for attackers. NEAR’s Chain Signatures takes a different route: the account that signs a transaction on another chain never holds a private key, because the key is never assembled in one place.
The mechanism is threshold multi-party computation. A single NEAR account can hold and move assets on other blockchains without a separate wallet or key for each one. The signing key is split into pieces held by 17 independent operators, and at least 11 must cooperate to produce a signature. No operator ever sees the whole key, and no single failure or compromise can move funds. NEAR’s documentation lists the networks the design reaches: Bitcoin, Solana, Cosmos, XRP, Aptos, Sui, and EVM chains including Ethereum, Base, BNB Chain, Avalanche, Polygon and Arbitrum.
On October 15, 2026, SVRN, a NASDAQ-listed company whose treasury holds more than 55 million NEAR tokens, announced it had finished migrating that operator network into tamper-resistant hardware. Every node now runs inside a trusted execution environment built on Intel TDX, a sealed enclave that the operator’s own administrators, its hosting provider and the operating system cannot read. The enclave also produces cryptographic proof of exactly which software is running, so anyone can check a node against the network’s published code rather than taking an operator’s word for it.
The threshold design is not risk-free, and it is worth being precise about what it buys. Eleven of seventeen is an honest-majority assumption: one bad or broken operator is harmless, but a coordinated majority could still reconstruct the key. That is why operators are vetted before they join the signing set, why the incentive program pays them monthly in arrears only for delivered uptime, and why adding or removing an operator, changing the threshold or updating node software requires an on-chain vote. Trail of Bits has audited the core components, but an audit records a review, not a guarantee, and trusted execution environments carry their own attack surface, including side channels.
What the design removes is the structural custodian. Because signatures are produced cooperatively inside attested enclaves, NEAR Intents, the settlement layer built on Chain Signatures, can settle transactions across networks without a bridge or custodian in the middle. SVRN reports that NEAR Intents has processed more than $34 billion in cross-network volume; that figure is the company’s accounting and should be read as such. The real change is in the failure modes: a bridge’s failure modes are replaced by a threshold network’s failure modes, and the migration’s value is that those new failure modes are visible, auditable and individually verifiable.
That visibility matters most as AI agents begin to transact on people’s behalf. An agent that must manage dozens of private keys is a liability; an agent that signs through a threshold network without ever possessing a complete key carries a different risk. The questions to watch are concrete: whether more operators adopt verifiable hardware, how settlement volumes respond after the migration, and whether other networks copy the incentive structure that SVRN’s HSP-007 proposal established. The migration does not settle whether a network can sign across chains. It sharpens the question of who holds the pieces, and what stops them from colluding.
Source: BTC-Pulse.