In 2026, Binance was investigating a security incident with a preliminary reported loss of roughly $6.3 billion. Tether and Circle said they froze at least $2.367 billion in dollar tokens tied to it—tokens that ordinarily move across public blockchains without either issuer approving each transfer.

The first dollar tokens already had operators

Stablecoins answered a practical question: how could someone move dollar-denominated value over open networks without waiting for a bank transfer? The transfer rail was new. The dollar claim still depended on an organization that issued tokens, managed access to redemption, and operated the service around them.

In 2017, Tether said hackers had stolen $30 million in tokens and suspended its back-end wallet service. It reopened verified customer deposits and fiat redemptions in 2018. A holder could see transfers on a chain, but getting dollars back through Tether's customer system required the operator to admit and serve that customer.

The intervention power was visible before the Binance episode, too. CENTRE blacklisted a USDC address holding $100,000 in 2020 after a law-enforcement request. In 2022, Tether froze three Ethereum addresses holding $160 million at law enforcement's request; blockchain data cited in that coverage counted 563 addresses frozen since 2017. Those cases established administrative control years before the Binance incident, even as the tokens traveled across networks as portable dollars.

Binance made crisis control large enough to measure

During the potential Binance incident, Tether CEO Paolo Ardoino said the company froze more than $1.567 billion in USDT. Circle CEO Jeremy Allaire said Circle froze more than $800 million in USDC in coordination with law enforcement. Binance's initial investigation pointed to a possibly compromised DevOps laptop that may have enabled unauthorized withdrawals; Binance said it had found no evidence of a wallet-private-key leak. Its reported loss of roughly $6.3 billion was still preliminary.

The issuers' stated freezes total at least $2.367 billion, a nominal 37.6% of that reported loss. That comparison shows the scale of issuer intervention, not the share of stolen property recovered. Binance's reported loss involved multiple assets, the investigation had not settled its final amount, and immobilized tokens had not been returned to their owners.

A blockchain can show where tokens went without deciding whether a valid transfer should be stopped. Issuers can make that decision for their own tokens once they identify an address they can reach. In early Binance reporting, on-chain data showed about $3.55 billion arriving at an unlabeled wallet and nearly $900 million subsequently moving across other wallets. Investigators had to identify balances while they were moving; Tether and Circle could not halt every asset in a multi-asset breach.

A freeze count needs a denominator—and a clock

Tether says it has frozen $4.2 billion linked to alleged illicit activity, including $3.5 billion since 2023. It also said in 2023 that it had proactively frozen USDT at addresses sanctioned by the U.S. Treasury's OFAC. Those amounts establish that intervention extends beyond one exchange incident. They do not establish how much value a freeze prevented from leaving, or how much anyone recovered.

For a particular crisis, an operator needs to know which balances belong to its token, which addresses it can identify in time, who can authorize a hold, and what happens after it acts. To judge the response, compare attributable exposure with the portion immobilized before it moves beyond the issuer's reach. A freeze count alone lacks that denominator and says nothing about the interval between detection and action.

Public scrutiny gives the question urgency without answering it. A Senate report found that 87% of 757 wallets it classified as terrorism-related predominantly transacted in USDT. That is a finding about the report's selected wallets, not about USDT holders generally. It helps explain why authorities examine issuer response procedures so closely—and why the criteria for identifying an address matter as much as the ability to stop it.

Transfer control also cannot repair a weak dollar claim. When Silicon Valley Bank failed in 2023, USDC lost its dollar peg and saw $1 billion in net redemptions after Circle disclosed that the bank held some of its cash reserves. Circle could have retained every address-level control and still faced a separate problem: whether holders trusted the assets and banking access behind redemption. A payments system needs both a credible claim and a way to contain disruption.

Institutions are buying the operator as well as the rail

Circle received U.S. regulatory approval to establish a national digital-currency trust bank and said the charter would enable institutional custody services. Paxos applied for a national trust bank charter after Circle and Ripple sought similar federal oversight. The Federal Reserve has proposed requiring some stablecoin issuers to back tokens fully with permissible reserve assets, including short-term Treasury bills. Those are different decisions—custody authority, a charter application, and a reserve proposal—but each puts an identifiable institution behind the token's dollar claim.

Distribution is tightening that relationship. In September, Binance acquired a $100 million stake in Circle and signed a five-year agreement to promote USDC. The deal does not explain Circle's later freeze or grant Binance authority over it. It does show how an exchange can depend commercially on a separately governed issuer: Binance supplies access to users; Circle remains responsible for the token it issues.

Tether has taken a different route into the U.S. market, launching USAT as a U.S.-regulated stablecoin issued by Anchorage Digital Bank alongside USDT. USAT's regulated U.S. issuance says nothing by itself about how USDT is governed.

More networks multiply the decisions an issuer must own

Stripe's Bridge launched Open Issuance so companies could create and manage their own stablecoins; Phantom used it to issue CASH. Visa said its stablecoin-settlement pilot supported nine networks and had reached a $7 billion annualized run rate. More issuers and routes give businesses more ways to move value. They also give counterparties more issuance terms, redemption paths, and escalation procedures to reconcile.

Tether said USDT would return to Bitcoin through Tether-backed Utexo, with most transaction data kept off Bitcoin's public ledger. That proposed arrangement separates a token's network destination from the visibility of its operating system. Meanwhile, Circle's president said the company was exploring reversible transactions for fraud or dispute refunds. Exploration is not an implemented right to reverse settled USDC transfers; it shows which payment-network decisions Circle is considering.

Issuer power has limits that ordinary payment failures expose. In 2025, Paxos said it mistakenly minted $300 trillion of PayPal's PYUSD on Ethereum and burned the tokens more than 20 minutes later. Burning an erroneous issuance corrected Paxos's own mistake; it did not demonstrate that Paxos could undo any completed payment between holders. The mint and the burn both remained in the ledger's history. In Binance's incident, the frozen tokens remained on the ledger, too; their holders could no longer send them.