How Did Hackers Steal $6.3 Billion From Binance? One Fake LinkedIn Job Interview and a Poisoned Download
Updated November 4, 2026 · Based on Binance’s report as covered by Fortune
North Korean hackers posed as recruiters on a job-hunting platform, ran two fake job interviews with a Binance engineer, and got the engineer to download a malicious “assignment.” A tampered software download promoted through an online ad did the rest. No deepfakes were reported: it was conventional phishing, which is exactly what a fake recruiter scam looks like.
How Did the North Korean Hackers Get Into Binance?
They used two separate tricks, both aimed at engineers who handled company wallets. Binance described the steps in a report released on Tuesday.
- A fake recruiter made contact. A person posing as a Project Director approached a Binance engineer through a legitimate job-hunting platform. The engineer’s profile mentioned experience at Binance.
- Two fake job interviews followed. The first was with someone posing as an HR representative. The second was with someone claiming to be the CEO of a fictitious company.
- An “assignment” carried the malware. The engineer was persuaded to download it, and that let the attackers into the company’s systems.
- A poisoned download added more. Another engineer clicked an ad promoting a tampered version of Homebrew, a popular software package, instead of using the company’s regular channels. Binance says it detected and removed the fake download, but the attackers had already learned who managed the company’s wallets.
Together, the two tricks let the attackers take over a series of wallets and drain them.
Did the Binance Hackers Use AI or Deepfakes?
Not for the break-in, according to Binance. Many people first guessed that AI had been used, but the company’s account points to conventional phishing.
AI played a role only after the attackers were inside: with an engineer’s laptop under their control, they used Binance’s own internal AI tool to take over accounts of other engineers.
Binance’s report doesn’t say the fake interviewers used deepfake faces or cloned voices. It matters anyway: the same recruitment trick can be paired with a synthetic face or voice on a video interview, which is why deepfake detection methods are worth knowing.
Is that interviewer a real person? How to check a video for AI
If a “recruiter” insists on a video call or sends you a clip, ask yourself: is this video AI? An AI video detector (also called an AI video checker) analyzes a recording for signs of synthetic generation or face swaps. The same check works on clips from Instagram reels and posts and Facebook videos, where fake recruiter and investment scams also spread. Voice notes can be faked too, so an AI voice detector is worth using on audio messages.
How Common Are Fake North Korean Recruiters?
Common enough that Binance says it had already flagged earlier attempts. The company says it detected and prevented 57 North Korea–related hacking attempts in 2026.
- Fortune reports that hackers posing as recruiters have become a widespread problem across the corporate sector.
- A Binance spokesperson said employees in non-public-facing roles are not allowed to keep LinkedIn profiles. The spokesperson didn’t say whether the targeted engineer broke that rule.
- Binance paused withdrawals for about 48 hours and re-enabled them on Saturday.
How to Spot a Fake Recruiter: 5 Red Flags
- The approach uses your employer. A message that cites your current company is built to look credible. Treat it as a target, not a compliment.
- You can’t verify the company or the interviewers. Check who they are through independent sources before the second round.
- A “test” or “assignment” needs a download. A real interview rarely asks you to run unknown files on a work computer.
- Software comes from an ad. Install tools only from official channels, never from a sponsored search result.
- You are on a video call with someone you can’t confirm. A synthetic face or voice is a risk, so confirm identity another way.
What if you think you’ve been targeted?
Don’t open the file. Stop talking to the recruiter, report the profile to the platform, and tell your security team if the contact reached you through a work device or work profile.
How Can You Protect Yourself From Fake Recruiters, Phishing Links and Deepfakes?
UncovAI wasn’t involved in this case, and Binance didn’t report deepfakes. But the tools match the tricks used, from a phishing link to a suspicious video call:
- Our AI scam and deepfake detector checks messages, images and media for signs of fraud and synthetic content.
- URL phishing detection checks a link before you click it, including links in ads and messages.
- Real-time deepfake detection for meetings flags synthetic faces and voices on video calls, such as a fake interviewer.
- Our AI video detector answers “is this video AI?” by checking a clip or a link, free to try.
- The deepfake Telegram bot lets you verify a video, photo or voice note without leaving Telegram.
- Our AI content detector covers deepfake, text and image detection in one place.
Comparing options? See the 10 best deepfake detection tools of 2026 and our guide to deepfake detection methods in 2026.
FAQ: The Binance Hack and Fake Recruiter Scams
How did hackers breach Binance in the $6.3 billion heist?
According to Binance, North Korean hackers contacted an engineer through a job-hunting platform, ran fake job interviews, and persuaded the engineer to download an assignment that let them into Binance’s systems. A tampered Homebrew download promoted through an ad gave them more information about wallet-management roles.
Did the Binance hackers use AI or deepfakes?
Binance says the attack relied mainly on conventional phishing. AI played a role only after the hackers controlled an engineer’s laptop: they used Binance’s own internal AI tool to take over other accounts. Binance did not report deepfakes.
How much was stolen and recovered?
Binance says hackers stole about $6.3 billion and that about $2.5 billion was recovered after Tether and Circle froze stolen stablecoins. Binance says all customer funds are safe and 98% of the stolen coins have been replenished.
What is a fake recruiter scam?
A fake recruiter scam is a phishing attack in which a criminal poses as a recruiter or hiring manager, often on LinkedIn, to build trust and then gets the target to download malware disguised as a test, assignment or software tool.
How can I spot a fake recruiter?
Verify the company and the people interviewing you independently, never download an assignment or test from a recruiter, and only install software from official sources, not from ads. An AI scam detector can check suspicious messages and links.
Can an AI scam detector or deepfake detector stop a fake job interview?
It can help. A URL phishing check screens links before you click, an AI scam detector reviews suspicious messages, and real-time deepfake detection can flag synthetic faces or voices on a video call. Binance did not report deepfakes in this case, so these tools complement, not replace, identity verification.
How do I check if a video or interview is AI-generated?
Upload the clip or paste its link into an AI video detector or AI video checker, such as UncovAI’s video detection tool. It looks for signs of synthetic generation or face swaps. Also confirm the person’s identity through a separate, trusted channel.
Is there a deepfake Telegram bot to verify a video, photo or voice note?
Yes. UncovAI offers a deepfake Telegram bot that verifies videos, photos and voice notes inside Telegram, so you can check suspicious media without switching apps.
What is UncovAI?
UncovAI is an AI content detector covering deepfake, text, image, video and voice detection. It also offers URL phishing detection and a free-to-start account; see the pricing page for plans.
Sources
- Fortune, “North Korea used LinkedIn, fake job interviews to trick Binance engineer in $6.3 billion hack,” by Jeff John Roberts, November 3, 2026, citing Binance’s report and statements to reporters.
Check before you click, download or reply
Run suspicious links, messages and calls through UncovAI’s AI scam detector and get a verdict in seconds.
Get started free →