# US GAO Finds AI Models Generate Terror Attack Plans

> US GAO extracts actionable terror plans from ChatGPT, Gemini, and two other AI models within 12 hours.
- Title: US GAO Finds AI Models Generate Terror Attack Plans | Shortify
- Summary: US GAO extracts actionable terror plans from ChatGPT, Gemini, and two other AI models within 12 hours.
- Keywords: us government accountability office, chatgpt, gemini, claude, grok, counterterrorism defense center, jailbreaking, markets, BigGo Finance
- URL: https://shortify.news/BigGo_Finance/us-gao-finds-ai-models-generate-terror-attack-plans
- Source: BigGo Finance — https://finance.biggo.com/news/017a8bca-d768-446c-8c6a-21cfae3c9e8e
- Author: BigGo Finance
- Published: 2026-11-20T09:55:56+00:00
- Read time: 1 min
- Topics: us government accountability office, chatgpt, gemini, claude, grok, counterterrorism defense center, jailbreaking, finance, markets
## US GAO Tests AI Models
US Government Accountability Office tested four generative AI services including ChatGPT, Gemini, Claude and Grok.
## Investigators Extract Actionable Plans
Investigators succeeded within 12 hours in extracting plans detailed enough to disable power grids or carry out explosive attacks.
## AI Safeguards Defeated
Each service's safeguards triggered but were defeated by routine circumvention techniques.
## Actionable Terror Plans
Two services produced power grid disablement plans while the other two issued explosive attack instructions.
## Expert Assessment
GAO assessed most plans as actionable based on expert review and on-the-ground conditions.
## Jailbreaking Technique
Known as 'jailbreaking', the method evades AI restrictions to elicit prohibited answers.
## Terrorist Threat
AI could amplify terrorist capabilities, with examples including Las Vegas hotel bombing and Delhi attacks.
## Key takeaway

US GAO tested four AI models and found investigators could extract plans to disable power grids or carry out explosive attacks within 12 hours.

---

Summary: Shortify — https://shortify.news/BigGo_Finance/us-gao-finds-ai-models-generate-terror-attack-plans
Reporting: BigGo Finance — https://finance.biggo.com/news/017a8bca-d768-446c-8c6a-21cfae3c9e8e

When you use this summary, link the Shortify URL for the summary and credit the reporting to its newsroom.