Deepfake Cyber Attacks: A Rising Threat in East Asia

In recent years, Advanced Persistent Threat (APT) groups in East Asia have been increasingly leveraging deepfake technology to execute sophisticated cyber attacks, posing critical risks to regional security. These attacks encompass deepfake video and audio social engineering, synthetic identity operations, AI-generated phishing media, and Business Email Compromise (BEC) fraud via deepfake voice calls.

Deepfake Video and Audio Social Engineering

APT groups are utilizing deepfake technology to create convincing video and audio content that impersonates trusted individuals, thereby deceiving targets into divulging sensitive information or performing unauthorized actions. For instance, in July 2025, the North Korean-linked Kimsuky group employed AI-generated deepfake images to impersonate officials from a South Korean defense-related institution, facilitating a spear-phishing attack. (ciso.economictimes.indiatimes.com)

Synthetic Identity Operations

By generating synthetic identities through deepfake technology, APT groups can infiltrate organizations undetected. The Chollima APT group, associated with North Korea, has been reported to use AI-powered facial filters during live video interviews to pose as legitimate job candidates, thereby gaining access to sensitive information within crypto and Web3 companies. (socradar.io)

AI-Generated Phishing Media

The integration of AI into phishing campaigns has led to the creation of highly personalized and convincing fraudulent communications. A 2026 report from Kaseya highlighted that 83% of phishing emails now incorporate AI, with 40% of BEC attacks utilizing generative AI. These AI-generated emails boast a 54% click rate due to improved grammar, personalization, and timely content, compared to just 12% for traditional phishing messages. (itpro.com)

BEC Fraud via Deepfake Voice Calls

Deepfake voice technology has been increasingly exploited in BEC attacks, where cybercriminals impersonate executives or trusted individuals to authorize fraudulent transactions. In Q4 2024, AI-generated deepfake fraud calls dominated scams, with consumers reporting losses exceeding $6,000 per incident. (businesswire.com)

Conclusion

The utilization of deepfake technology by APT groups in East Asia represents a significant escalation in cyber threat sophistication. Organizations must enhance their cybersecurity measures, including AI-driven detection systems, employee training on recognizing deepfake content, and robust verification protocols for sensitive communications, to mitigate the risks associated with these advanced cyber attacks.