AUTO-UPDATED

‘Just ask Grok’: How ISIL is using Big Tech’s AI to build bombs

Boko Haram fighters are increasingly utilizing major AI chatbots to manufacture explosives, plan combat operations, and circumvent safety protocols, according to new research from the University of Cambridge.

Key Points

  • Former members of ISWAP and JAS reported using ChatGPT, Claude, Gemini, Grok, Meta AI, and DeepSeek for technical and military guidance.
  • Dedicated AI units within the militant groups used laptops and projectors to train commanders on bypassing safety guardrails.
  • Fighters consulted AI to optimize bomb-making processes, repurpose unexploded military ordnance, and adapt commercial drones for weaponized use.
  • Pro-ISIL forums are actively sharing methods to "jailbreak" AI models to remove restrictions on generating dangerous or weapons-related information.
  • The study highlights that AI provides real-time, tailored technical expertise that functions as a 24-hour consultant for battlefield challenges.

Why it Matters

The integration of advanced AI into militant operations demonstrates how rapidly emerging technologies can be exploited to lower the barrier for creating sophisticated weaponry and tactical planning. This trend challenges the effectiveness of current industry safety guardrails and intensifies the debate over whether AI development requires stricter global regulation to prevent misuse by armed groups.
Al Jazeera English Published by Caolán Magee
Read original