A University of Cambridge study found that Boko Haram fighters used AI chatbots from major technology companies, including ChatGPT, Claude, Gemini, Grok, Meta AI and DeepSeek, to obtain technical and military information such as advice linked to explosives, weapons and combat operations. Researchers interviewed 27 former members of ISWAP and JAS; several said they used the tools before, during and after attacks, and one former ISWAP commander said, “There is nothing we haven’t received an answer on.” Members were trained to use AI systems and circumvent safeguards, and the groups created specialised AI units of engineers, weapons experts and bomb-makers, including a 23-member unit operating from a solar-powered room. Study author Antonia Juelich told Al Jazeera that about 10 of the 27 interviewees spoke about using AI in connection with explosives. Fighters reportedly sought information on recovering explosive materials from unexploded military ordnance and modifying commercially available drones to carry explosives, used AI during combat, including uploading battlefield images to ChatGPT for analysis, and examined failed attacks. Separately, Tech Against Terrorism found discussions among supporters of the wider ISIL movement about bypassing AI safety controls and obtaining weapons-related information; Lucas Webber warned the material uncovered could represent only part of the activity.