North Korean hackers use AI to forge military IDs
South Korean cybersecurity firm Genians says the North Korean hacking group Kimsuky used ChatGPT to generate a realistic draft of a South Korean military ID and included the forged document in phishing emails impersonating a credentialing body. The campaign — disclosed in a Genians blog post and reported Sept. 26, 2025 — illustrates how generative AI lowers the barrier for sophisticated cyberespionage; the article also cites OpenAI and Anthropic disclosures that Chinese and other state-linked hackers have used chatbots to probe U.S. defense and infrastructure systems.
AI & Tech
National security
📰 Sources (1)
North Korean hackers use AI to forge military IDs
New information:
- Genians discovered Kimsuky used ChatGPT to produce a fake draft South Korean military ID which was attached to phishing emails targeting military-affiliated officials.
- The attackers bypassed built-in chatbot safeguards by framing prompts as 'sample designs for legitimate purposes,' producing realistic-looking mockups.
- The report links to broader patterns: OpenAI and Anthropic say other state-linked hackers used ChatGPT and Claude to assist attacks, including reconnaissance on U.S. defense networks and satellite systems.