"God Helped Us, So Will AI": How Boko Haram Used ChatGPT, Grok For Attacks

The use of artificial intelligence is no longer just limited to content generation and it being used as a chatbot to answer questions. Terrorist group Boko Haram and its factions have been using artificial intelligence tools from major technology companies to build explosives, plan attacks and solve problems during combat, according to a new study by the University of Cambridge.

The findings are based on interviews with 27 former members of Boko Haram's two factions - the Islamic State affiliate in West Africa Province (ISWAP) and Jama'at Ahl as-Sunnah lid-Da'wah wa'l-Jihad (JAS), Al Jazeera reported.

The former terrorists told researchers that they had used ChatGPT, Claude, Gemini, Grok, Meta AI and DeepSeek for a range of military activities.

Boko Haram Members Trained To Use AI

According to the research, the use of AI within the groups was not limited to individual terrorists experimenting with chatbots. Former terrorists described organised training sessions in ISWAP-controlled territory in northeastern Nigeria, where senior operatives were taught how to use AI systems.

Terrorists were brought into rooms with laptops and projectors, where they were shown how to interact with AI chatbots and get around their safety restrictions.

"The white guys came and taught us," one former commander recalled. "They assembled the top people in a room and used a projector to show how it works on a big screen."

The trainers also provided laptops equipped with privacy and encryption tools and arranged accounts and paid subscriptions for several AI platforms, according to former members cited in the study.

'How Can I Build A Bomb?'

The research found that ISWAP and JAS went on to establish dedicated AI units, with engineers, weapons specialists and bomb-makers among their members.

One former JAS commander described a 23-member unit operating from a solar-powered room. Members answered questions, analysed information and trained senior commanders who gathered around a laptop.

AI was used by bomb-makers during the process of assembling explosives, with terrorists returning to chatbots whenever they encountered a problem.

"'[Ask] how can I build a bomb?' and then it tells you how," explained a former Boko Haram commander. "It is like a human robot. We used it a lot."

Another former member said, "God has helped us, and so will AI."

Antonia Juelich, who authored the Cambridge study, told Al Jazeera that the terrorists could provide chatbots with details of the materials they had and ask follow-up questions when they encountered problems.

This meant the technology could provide responses tailored to the materials available to them rather than simply giving generic information, she said.

AI was also used to modify commercially available drones to carry and drop explosives on targets.

ChatGPT Used To Analyse Battlefield Footage

In one instance described by a former terrorist, a terrorist entered the battlefield with a camera attached to his chest and transmitted images back to the group's base.

A commander uploaded the footage to ChatGPT, used the chatbot to analyse the situation and then relayed fresh instructions to terrorists on the frontline.

AI was also consulted when the groups obtained unfamiliar weapons.

In one case, commanders asked an AI specialist how to operate newly acquired firearms. His response was simple, "Just ask Grok," referring to billionaire Elon Musk's AI chatbot.

AI Used To Analyse Failed Attacks

The technology was even used after operations ended.

Former members told researchers that terrorists would return from unsuccessful missions and describe the tactics they had used to AI systems, asking why the operation had failed.

Footage from operations was also uploaded for analysis, with terrorists using the responses to change their tactics for subsequent attacks.

The Cambridge interviews mainly covered AI use during 2023 and 2024. Juelich told Al Jazeera that the findings were particularly concerning because AI systems have developed rapidly since the period covered by the interviews.

The revelations come amid a broader debate over safeguards for increasingly powerful AI systems.

Anthropic chief executive Dario Amodei has called for AI companies to slow the development of their most advanced models to give safety measures time to catch up. He has also backed independent evaluations and common safety standards.

However, US President Donald Trump has taken a different approach, arguing that American AI companies should be allowed to innovate without "cumbersome regulation" and pursuing what it calls a "minimally burdensome" national framework.