Cambridge Study Finds Boko Haram and ISWAP Used AI Chatbots for Attack Planning
A University of Cambridge field study, based on interviews with former Boko Haram and ISWAP members in north-east Nigeria, found that both armed groups established dedicated internal units to use commercial AI chatbots such as ChatGPT, Claude, and Gemini. The groups reportedly applied these tools to attack planning, weapons troubleshooting, and explosive design, often bypassing safety guardrails through disguised prompts. Organizational adoption is believed to have begun around 2023, with access tightly restricted to senior members and some units receiving training from foreign fighters. The findings highlight a gap in mainstream AI safety frameworks, which have largely focused on state-level threats and broad societal risks while overlooking resourceful non-state armed groups. Researchers are now calling for AI safety evaluations to include conflict and terrorism experts, and for behavioral detection systems capable of identifying concentrated, iterative misuse.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in