AI Incident Database: Underground Market for LLMs Powers Malware and Phishing Scams

Event summary

A study by Indiana University researchers uncovered widespread misuse of large language models (LLMs) for cybercrime. Cybercriminals, according to that study, use LLMs like OpenAI's GPT-3.5 and GPT-4 to create malware, phishing scams, and scam websites. These models are available on underground markets, often bypassing safety checks through jailbreaking. Named malicious LLMs are BadGPT, XXXGPT, Evil-GPT, WormGPT, FraudGPT, BLACKHATGPT, EscapeGPT, DarkGPT, and WolfGPT.

Linked entities

• OpenAI

  vendor | 70%

• cybercriminals

  organisation | 70%

• badgpt

  organisation | 70%

• xxxgpt

  organisation | 70%

• evil-gpt

  organisation | 70%

• wormgpt

  organisation | 70%

• fraudgpt

  organisation | 70%

• blackhatgpt

  organisation | 70%

• escapegpt

  organisation | 70%

• darkgpt

  organisation | 70%

• wolfgpt

  organisation | 70%

• internet-users

  organisation | 70%

• organizations

  organisation | 70%

• individuals-targeted-by-malware

  organisation | 70%

Related graph edges