Binance CSO Described Fake Job Attack Before $6.3B Hack

Binance CSO Described Fake Job Attack Before $6.3B Hack

Binance Chief Security Officer Jimmy Su described a fake recruitment scenario months before attackers used a similar technique as part of the exchange’s $6.3 billion security breach. But the detailed warning was not publicly available before the attack: remarks Su made during a July interview about malicious coding assessments were only published on November 12, after the October 29 theft. Binance’s official preliminary investigation has since confirmed that a fake job interview formed one stage of the intrusion.

Fake Recruitment Became an Entry Point

During the July interview, Su explained that attackers could approach employees with tailored job opportunities and ultimately persuade them to execute malicious code through a technical assessment. The published July coverage did disclose that Binance internally simulated recruiter-based phishing attacks, but the more specific coding-assessment scenario that closely resembles the later breach was not included at the time.

That timing materially changes the security interpretation. There is no evidence that the attackers learned their technique from Su’s comments or that Binance’s public threat disclosures reduced the work required to identify the attack path. The similarity shows that Binance security staff understood fake recruitment as a serious social-engineering threat before the breach, not that their disclosure caused or facilitated it.

Binance says the actual operation developed over several months. A separate engineer installed a compromised software package in June, while another employee was targeted through a fake recruitment process in September and downloaded a malicious coding assignment onto a work computer. By October 20, the attackers had used compromised infrastructure and an internal AI tool to reach additional wallet engineers, according to Binance’s subsequent reconstruction.

The final theft required considerably more than control of one laptop. Attackers ultimately gained unauthorized access to an internal wallet transaction-signing service, allowing transfers outside Binance’s normal withdrawal process. On Ethereum, they used EIP-7702 delegation to grant attacker-controlled code execution rights over a hot wallet. The breach therefore combined human compromise, credential access, internal infrastructure and wallet authorization mechanisms rather than exploiting a single disclosed vulnerability.

That multichain compromise produced approximately $6.34 billion in affected assets, including BTC, XRP, SOL, OP and Ethereum-based tokens. The mechanics of the $6.3 billion Binance breach and subsequent asset freezes show why the EIP-7702 component explains only part of the incident: unauthorized transactions also occurred on networks where that Ethereum feature does not exist.

Known Threat Did Not Prevent the Breach

The episode exposes a different security problem from the one suggested by the original disclosure narrative. Binance was already actively defending against fake recruitment and phishing attacks before one succeeded. Su said the exchange had run monthly employee phishing simulations for several years, including scenarios where its internal red team posed as recruiters.

The attackers nevertheless invested enough time to bypass those controls. The FBI formally attributed the theft to North Korea’s TraderTraitor operation on October 31, while Binance said it had blocked 57 suspected North Korea-linked attempts during 2026. Stopping repeated attacks does not eliminate the possibility that one successful employee compromise can become the first stage of a much deeper intrusion.

Containment also demonstrated the importance of controls beyond endpoint security. The malicious EIP-7702 authorization remained effective independently of Binance’s normal withdrawal infrastructure, so suspending withdrawals could not immediately stop every transfer. Binance ultimately had to revoke the authorization onchain, while its MPC wallet architecture meant there was no single private key capable of instantly overriding it. The incident turned an employee-device compromise into a wallet-control problem because several security boundaries failed sequentially.

Binance has since rebuilt the affected wallet environment, replaced compromised devices and credentials, strengthened controls around its internal AI systems and commissioned an independent cybersecurity review. It also used company resources and its user-protection reserve to replenish affected assets, including a substantial drawdown of its SAFU Bitcoin holdings. As of November 3, Binance said 98% of affected token assets had been replenished and approximately $2.85 billion had been frozen or intercepted.

The sharper lesson is not that Binance disclosed too much about an attack vector, but that recognizing a social-engineering technique is different from preventing a determined adversary from executing it. The July remarks show that the threat was already understood internally; the October breach shows how a successful foothold can still cascade through credentials, development infrastructure and wallet systems when an attacker has enough time to move laterally.

Follow Us

Ads

Main Title

Sub Title

It is a long established fact that a reader will be distracted by the readable

Ads
banner 900px x 170px