Skip to Content
logologo
AI Incident Database
Open TwitterOpen RSS FeedOpen FacebookOpen LinkedInOpen GitHub
Open Menu
発見する
投稿する
  • ようこそAIIDへ
  • インシデントを発見
  • 空間ビュー
  • テーブル表示
  • リスト表示
  • 組織
  • 分類法
  • インシデントレポートを投稿
  • 投稿ランキング
  • ブログ
  • AIニュースダイジェスト
  • リスクチェックリスト
  • おまかせ表示
  • サインアップ
閉じる
発見する
投稿する
  • ようこそAIIDへ
  • インシデントを発見
  • 空間ビュー
  • テーブル表示
  • リスト表示
  • 組織
  • 分類法
  • インシデントレポートを投稿
  • 投稿ランキング
  • ブログ
  • AIニュースダイジェスト
  • リスクチェックリスト
  • おまかせ表示
  • サインアップ
閉じる

レポート 3643

関連インシデント

インシデント 63430 Report
Alleged Deepfake CFO Scam Reportedly Costs Multinational Engineering Firm Arup $25 Million

Loading...
Employee Duped by AI-Generated CFO in $25.6M Deepfake Scam
hackread.com · 2024

Learn how a multinational company's Hong Kong branch faced a deepfake scam, causing substantial financial losses, and explore efforts to counter the rising sophistication of deepfake technology.

A multinational company's Hong Kong branch faced a substantial financial setback due to an advanced deepfake scam. During a video call, an employee was deceived by digitally manipulated versions of the company's CFO and others, resulting in the unauthorized transfer of funds to the scammers.

The scam recalls the August 2022 incident in which scammers created a Deepfake AI hologram of Patrick Hillmann, Binance's Chief Communications Officer. The hologram was manipulated to deceive users into participating in online meetings and to target crypto projects of Binance clients.

Reports indicate that the employee was tricked into paying out HK$200 million (approximately US$25.6 million) after participating in a deepfake video call impersonating the company's CFO. The worker was persuaded after seeing multiple staff members attending that call. However, all were deepfake recreations. 

The worker initially became suspicious after receiving a message from the UK-based CFO and avoided it believing it was a phishing email. However, after a video call, the worker realized that the other attendees resembled his colleagues, despite initial doubts, as they appeared and sounded familiar.

What happened is that the scammers used publicly available footage to create convincing versions of the meeting's participants, according to Hong Kong police senior superintendent Baron Chan Shun-ching's briefing on Friday. 

In the digitally altered video, the CFO issued fraudulent instructions to transfer funds to unauthorized accounts. The employee, a finance worker at the company, followed the instructions, leading to a significant financial loss because the scammers had immaculately replicated the CFO's appearance and voice.

The worker transferred $25 million to five bank accounts in 15 transactions. The scam was discovered after the employee shared the information with the company's head office. It is worth noting that the incident lasted a week. 

However, this isn't the first of its kind incident involving deepfake technology. Fraudsters have been relying on this technology to cheat people out of money.

Bitdefender's latest research reveals a rise in YouTube stream-jacking campaigns using deepfake videos for cryptocurrency theft. Despite McAfee's MockingBird tool detecting 90% of deepfake content, scammers continue to use malicious techniques for crypto scams, sometimes bypassing facial recognition systems, highlighting ongoing challenges in combating these deceptive practices.

Authorities are investigating the incident in Hong Kong, which is the first such scam involving a large sum of lost money. It has raised concerns about the scammers' increasing sophistication and potential for financial fraud. Experts urge increased awareness and training for employees to effectively identify and resist these scams.

Hong Kong police confirmed that eight stolen identity cards were used for 90 loan applications and 54 bank account registrations between July and September 2023 using AI deepfakes to trick facial recognition programs on at least 20 occasions. The authorities have so far arrested six individuals in connection with deepfake scams.

情報源を読む

リサーチ

  • “AIインシデント”の定義
  • “AIインシデントレスポンス”の定義
  • データベースのロードマップ
  • 関連研究
  • 全データベースのダウンロード

プロジェクトとコミュニティ

  • AIIDについて
  • コンタクトとフォロー
  • アプリと要約
  • エディタのためのガイド

インシデント

  • 全インシデントの一覧
  • フラグの立ったインシデント
  • 登録待ち一覧
  • クラスごとの表示
  • 分類法

2024 - AI Incident Database

  • 利用規約
  • プライバシーポリシー
  • Open twitterOpen githubOpen rssOpen facebookOpen linkedin
  • e1b50cd