A new report has revealed that members of Boko Haram and the Islamic State West Africa Province (ISWAP) allegedly exploited artificial intelligence (AI) chatbots, including ChatGPT, Claude, Gemini, Grok and DeepSeek, to improve bomb-making techniques and overcome tactical challenges during terrorist operations.
Former Boko Haram and ISWAP commanders told researchers that the terrorist groups increasingly turned to artificial intelligence tools to enhance their operational capabilities, according to a report by The New York Times.
The report, which cited research conducted by Antonia Juelich of the University of Cambridge, said former insurgents disclosed how AI platforms were used to obtain solutions to battlefield challenges and generate technical guidance for attacks.
One former Boko Haram commander recounted how the group sought AI assistance after an attack on a military base was unsuccessful because a defensive trench prevented fighters on motorcycles from advancing.
According to the report, the insurgents asked AI chatbots how to modify their motorcycles to clear the obstacle, providing details about the type of motorcycles they used and the distance they needed to jump.
“We saw in a movie how motorcycles can jump over bridges. We used AI to learn how to do this,” the former commander was quoted as saying.
The AI platforms reportedly generated step-by-step instructions, enabling the group’s mechanics to improve the motorcycles’ speed and acceleration before fighters practised the manoeuvre.
The report also stated that AI chatbots were allegedly used to obtain detailed guidance on constructing explosive devices.
A former ISWAP commander reportedly said: “You type in the question or use your voice, and it gives you a detailed answer, like ‘How can I build a bomb?’, and then it tells you how. It is like a human robot. We used it a lot.”
According to the findings, members of the terrorist groups believed AI reduced the risks associated with trial-and-error methods, with one insurgent claiming that “AI gives you accuracy.”
Another former fighter reportedly told researchers that AI-generated guidance helped the group alter chemical compositions, making their explosive devices more powerful.
Reacting to the report, OpenAI said the use of its technology for terrorist activities violates the company’s policies.
Google and Anthropic also maintained that their AI systems are designed to reject requests involving weapons, explosives and other harmful activities.
However, the report noted that former insurgents claimed they were sometimes able to circumvent chatbot safety measures by disguising dangerous requests as educational or legitimate research projects.
