---
format: "aidr-story-markdown/v1"
id: "5a91e99e9146b9533392a7e359e602395b5e3bd32decad6a736772e9ea26913b"
canonical_url: "https://aidr.today/5a91e99e?lang=en"
title: "AI Models Can Be Easily Modified to Assist Terrorist and Extremist Activity"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-14T09:00:00.000Z"
category: "Research"
topics: ["safety","regulation","llm","national-security"]
source_urls: ["https://www.wired.com/story/ai-models-can-be-easily-modified-to-assist-terrorist-and-extremist-activity/","https://www.techmeme.com/261014/p16"]
summary: "Researchers demonstrated that AI models with removed guardrails can provide detailed instructions for creating improvised explosive devices when prompted with extremist intent."
---

# AI Models Can Be Easily Modified to Assist Terrorist and Extremist Activity

> [Open the canonical story](<https://aidr.today/5a91e99e?lang=en>)

**Published:** 2026-10-14T09:00:00.000Z
**Category:** Research
**Topics:** safety, regulation, llm, national\-security

## Summary

Researchers demonstrated that AI models with removed guardrails can provide detailed instructions for creating improvised explosive devices when prompted with extremist intent\.

## Sources

- [Story source](<https://www.wired.com/story/ai-models-can-be-easily-modified-to-assist-terrorist-and-extremist-activity/>)
- [Story source](<https://www.techmeme.com/261014/p16>)