Binance says $6.3 billion hit in largest crypto theft on record
The world's largest crypto exchange says it will cover losses from a breach that froze or stole an estimated $6.3 billion in client assets, as stablecoin issuers scramble to lock down affected funds.
Binance, the world's largest cryptocurrency exchange, disclosed this week that a security breach affected an estimated $6.3 billion in client assets, with roughly $3 billion frozen on-chain — what multiple outlets described as the largest crypto theft on record. Binance said it is covering losses through its Secure Asset Fund for Users (SAFU) and temporarily suspended withdrawals while overhauling its security, with plans to resume within 48 hours, according to a detailed CNBC report and a separate TechCrunch account.
How it happened
Binance initially said the breach may have stemmed from a compromised laptop belonging to a DevOps engineer, which let an attacker reach an internal service container and submit unauthorized withdrawals. Blockchain-intelligence firms moved quickly to assign blame: TRM Labs flagged North Korean hackers as likely culprits, citing similarities to the 2025 Bybit hack, while Chainalysis attributed the attack to North Korea's Lazarus-linked TraderTraitor group. The FBI said it is aware of the incident and is working with partners to investigate.
Stablecoin issuers race to freeze funds
Three major stablecoin issuers said they moved to lock down funds tied to the breach. Circle said it froze more than $800 million in USD Coin; Tether said it froze more than $1.5 billion of its token; and Ripple said it was actively freezing affected funds in its RLUSD stablecoin. The breach has quickly become a flashpoint in Washington's debate over crypto regulation, with CFTC Chairman Michael Selig saying the incident "proves our exact point" about bringing offshore exchanges under federal oversight.

Meta closes $35 billion stock sale for AI buildout as shareholders sue over Zuckerberg's new share class

UK competition regulator opens formal investigation into Meta over AI data-scraping terms
