A groundbreaking study from the University of Cambridge has uncovered how Boko Haram is weaponizing artificial intelligence to escalate its deadly campaign. The extremist group is actively using six major AI platforms—originating from both American and Chinese tech ecosystems—to refine attack strategies, engineer explosives, and streamline military operations.
How Boko Haram turned AI into a terror tool
Researchers at Cambridge’s AI Science & Policy Programme conducted 57 in-person interviews with former Boko Haram members, mid-level commanders, and technical experts. Their findings, covering activities from 2023 to mid-2025, reveal a dramatic shift: what began as propaganda tools have evolved into full-fledged instruments of war.
The study highlights the formation of dedicated AI units within Boko Haram, each equipped with subscriptions to multiple AI services. These cells respond to real-time requests from field operatives, providing tactical insights, bomb-making guides, and operational optimizations. The platforms in use—ChatGPT, Claude, Gemini, Grok, Meta AI, and DeepSeek—span the technological and geopolitical divide, from Silicon Valley to Beijing.
The fragmentation of AI security: a terror loophole
A critical vulnerability emerges from the lack of coordination between American and Chinese AI developers. With no shared protocols to flag or block malicious use, Boko Haram seamlessly switches between platforms, exploiting gaps in moderation policies. A Tech Against Terrorism-backed experiment tested 27 AI models with 2,300 terror-related queries. Shockingly, 32% yielded usable intelligence, rising to 42% when questions were rephrased with explicit intent.
This regulatory vacuum is exacerbated by U.S.-China geopolitical tensions. Each tech giant develops its own safeguards independently, creating blind spots that organized groups like Boko Haram exploit. The result? A fragmented AI landscape where terror groups operate with near impunity.
DeepSeek’s role: China’s AI advantage in the jihadist arsenal
The inclusion of DeepSeek—a Chinese AI platform—marks a turning point. Less scrutinized by Western authorities, it provides an alternative route when U.S.-based AI tools tighten restrictions. Terrorists now alternate between platforms, leveraging differences in moderation to bypass blocks. Training sessions, led by Islamic State-affiliated experts, teach operatives to bypass safeguards through strategic question reformulation.
The impact is chilling. AI has enabled Boko Haram to slash the number of fighters per operation from 200 to just 20, while boosting attack precision. Tactical analyses, escape route planning, and logistical optimizations—once reliant on trial and error—are now AI-generated, reducing risks for the group.
Sovereignty and security at stake
DeepSeek’s involvement raises urgent questions about digital sovereignty. China’s parallel AI ecosystem operates outside Western regulatory reach, complicating intelligence interception for European and U.S. agencies. The decentralized nature of these tools allows jihadist groups to access cutting-edge capabilities without centralized oversight.
By 2025, the threat had transcended Africa. Documented incidents surged in the United States, Canada, Israel, Finland, France, and Austria, where terrorists used AI to plan attacks. The cross-border diffusion of this expertise poses a direct challenge to Western security architectures.