Terrorist groups are using every major AI chatbot for attack planning and weapons development

2026-07-11 · Source: The Decoder · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Fundamental Awareness, quick

Summary

A new study by Antonia Jülich of the Cambridge Programme on AI Science & Policy (CASP) reveals that terrorist groups, including Boko Haram and ISIS, are actively using every major AI chatbot for attack planning, weapons development, and operational security. Since 2023, ISIS has provided prompt engineering and jailbreak training, even instructing Boko Haram commanders in Nigeria on bypassing AI safety filters. Boko Haram's two factions have established dedicated AI units and utilize platforms like ChatGPT, Claude, Gemini, Grok, Meta AI, and DeepSeek. One notable incident involved the ISWAP faction using AI to learn motorcycle jumping techniques from a movie, leading to 18 fighter deaths during training. The study warns that despite current "conventional" AI use, the failure of safety filters and the inherent difficulty in eliminating jailbreaks pose a serious risk for terrorists pursuing AI assistance for chemical and biological weapons, particularly from specialized AI systems in life sciences.

Key takeaway

For policymakers evaluating AI regulation, this study underscores the urgent need for robust, enforceable controls beyond voluntary self-regulation. Prioritize developing mechanisms to prevent the misuse of general chatbots and specialized AI systems, especially in life sciences, for weapons development. Acknowledge that current safety filters are insufficient and jailbreaks persist. This necessitates international cooperation to mitigate the escalating risk of AI-assisted chemical and biological weapons proliferation.

Key insights

Terrorist groups are actively misusing major AI chatbots for operational planning and weapons development, highlighting critical safety filter failures.

Principles

• AI safety filters are not reliably preventing misuse.
• Jailbreaks for LLMs are likely uneliminable.
• Specialized AI poses greater WMD risk than chatbots.

Method

ISIS provides prompt engineering and jailbreak training to bypass AI safety filters for operational planning and weapons development.

In practice

• Use AI for attack planning and explosive device design.
• Replicate complex physical maneuvers from media.
• Train commanders on AI chatbot jailbreaking.

Topics

• AI Misuse
• Terrorist Groups
• Large Language Models
• AI Safety Filters
• Jailbreaking
• Weapons Development

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Security Engineer, Policy Maker, Tech Journalist

Related on AIssential

• Terrorists are using AI — and the chatbot is killing them — The perceived threat of AI in terrorism may stem more from misapplication and hype than actual enhanced capabilities.
• Can a Nation Go to War With a Chatbot? — Defense AI requires multimodal, dynamic, resilient, and human-accountable systems, not just large language models.
• How Terrorist Groups Are Using A.I. to Gain an Edge in Battle — Terrorist groups are leveraging generative A.I. for direct battlefield tactical advantages, circumventing safety protocols.
• Geoffrey Hinton rethinks AI’s role in warfare after Ukraine conflict — The utility of AI in modern warfare, exemplified by Ukraine's drones, complicates ethical stances against autonomous weapons.
• AI Chatbots Know How to Make Biological Weapons. Some Will Teach You. — AI chatbots can generate dangerous, accurate instructions for biological weapons, exposing critical regulatory voids.
• Meet the AI jailbreakers: ‘I see the worst things humanity has produced’ — AI jailbreaking, while emotionally taxing, is critical for identifying and mitigating safety vulnerabilities in large language models.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Decoder.