# Anthropic report: Claude used for missile dev and influence ops in Middle East

Digest AI · Policy & Regulation · published 2026-09-10T00:57:00Z · updated 2026-09-16T10:21:00Z

Canonical: https://digestai.news/story/anthropic-reveals-fourth-claude-ai-security-breach

## Summary

Anthropic has released an intelligence report detailing how its Claude AI model was exploited by state and non-state actors across the Middle East between December 2025 and August 2026. The findings reveal that a cell in northern Yemen utilized Claude to develop guidance, navigation, and control software for ballistic missiles, including a multi-stage system with a range goal exceeding 2,000 kilometers. Although the group managed multiple AI instances to mimic an engineering team, Anthropic noted that their test-fire of a guided rocket failed, and no operational weapon was successfully fielded.

Beyond weapons development, the report highlights extensive malicious use in Iran and the United Arab Emirates. Iranian-linked actors used the model to compile targeting handbooks on US Navy vessels and execute propaganda campaigns through state-aligned institutions. Meanwhile, UAE officials allegedly deployed Claude to run a large-scale influence operation involving 300 fake social media accounts, targeting the Muslim Brotherhood and countering criticism of Abu Dhabi’s role in the Sudan conflict.

Anthropic stated that it banned the suspicious accounts and shared threat intelligence with government partners. The report underscores the growing risk of commercial AI models being integrated into active military and covert influence programs, particularly against the backdrop of the ongoing US-Israel-Iran conflict.

## Key points

- Yemeni actors used Claude to design missile guidance software, though a test-fire reportedly failed.
- Iranian entities used the model to create targeting handbooks for US Navy ships and run propaganda.
- UAE officials allegedly used Claude to manage an influence operation with 300 fake social media accounts.

## Why it matters

This report provides rare documented evidence of commercial LLMs being used in active weapons development and state-sponsored influence operations, highlighting critical gaps in AI safety safeguards against sophisticated adversarial evasion.

## Sources

1. [Claude AI used for missile, influence projects in UAE, Iran, Yemen: Anthropic](https://al-monitor.com/originals/2026/09/claude-ai-used-missile-influence-projects-uae-iran-yemen-anthropic) (al-monitor.com, 2026-09-11)
2. [AI makes it "easier than ever" for adversaries to target U.S. military, experts say](https://cbsnews.com/news/ai-us-adversaries-target-military-operations) (cbsnews.com, 2026-09-16)
3. [Russian team ‘misused’ Claude AI to train a killer drone on scraped Ukrainian war footage](https://theconversation.com/russian-team-misused-claude-ai-to-train-a-killer-drone-on-scraped-ukrainian-war-footage-291965) (theconversation.com, 2026-09-15)
4. [Claude AI used for weapons, cyber ops, surveillance, fraud in China and beyond: Anthropic](https://cnbctv18.com/technology/claude-ai-used-for-weapons-cyber-ops-surveillance-fraud-in-china-and-beyond-anthropic-19990504.htm) (cnbctv18.com, 2026-09-14)
5. [Iranian Operatives Using AI to Target Jews, Israelis, Anthropic Report Says](https://algemeiner.com/2026/09/14/iranian-operatives-using-ai-target-jews-israelis-anthropic-report-says) (algemeiner.com, 2026-09-14)
6. [Chinese state-linked actors used AI assistant Claude to target dissidents, religious groups](https://tribuneindia.com/news/world/chinese-state-linked-actors-used-ai-assistant-claude-to-target-dissidents-religious-groups) (tribuneindia.com, 2026-09-14)
7. [Houthis Used Claude Code to Develop Missile Guidance Software: Anthropic](https://clashreport.com/world/articles/houthis-used-claude-code-to-develop-missile-guidance-software-anthropic-s52mnx4pwpo) (clashreport.com, 2026-09-13)
8. [Chinese military researchers and tech giants caught using Claude — US frontier model, coded 16 air-defense suppression tools targeting Taiwan, drafted anti-torpedo specs, and fed 151 million training queries to Alibaba](https://tomshardware.com/tech-industry/artificial-intelligence/chinese-military-researchers-and-tech-giants-caught-using-claude-us-frontier-model-coded-16-air-defense-suppression-tools-targeting-taiwan-drafted-anti-torpedo-specs-and-fed-151-million-training-queries-to-alibaba) (Tom's Hardware, 2026-09-13)
9. [Anthropic CONFIRMS Claude AI Used by Yemeni Houthi to Build Missile Program Links To Iran Operation](https://oneindia.com/videos/anthropic-confirms-claude-ai-used-by-yemeni-houthi-to-build-missile-program-links-to-iran-operation-4323812.html) (oneindia.com, 2026-09-12)
10. [Iran and Houthi rebels used Anthropic's Claude AI to target US warships and build hypersonic missiles — Houthi rebels also used the bot to code ballistic missile guidance systems](https://tomshardware.com/tech-industry/artificial-intelligence/iran-and-houthi-rebels-used-anthropics-claude-ai-to-target-us-warships-and-build-hypersonic-missiles-houthi-rebels-also-used-the-bot-to-code-ballistic-missile-guidance-systems) (Tom's Hardware, 2026-09-12)
11. [Anthropic says Iran-linked group used Claude to track U.S. Navy ships](https://interestingengineering.com/military/anthropic-claude-iran-us-navy-ai-military-targeting) (interestingengineering.com, 2026-09-12)
12. [Anthropic reports Russian developers used AI for kamikaze drone software](https://cryptobriefing.com/anthropic-russian-ai-kamikaze-drone-software) (cryptobriefing.com, 2026-09-12)
13. [Anthropic report: 5 ways Claude was exploited for war, spying and repression](https://yahoo.com/news/politics/articles/anthropic-report-5-ways-claude-123602565.html) (yahoo.com, 2026-09-12)
14. [Did Yemen’s Houthis use Anthropic’s Claude AI To design Missiles?](https://newsx.com/world/did-yemens-houthis-use-anthropics-claude-ai-to-design-missiles-271072) (newsx.com, 2026-09-12)
15. [Anthropic says Iran, Russia used Claude for weapons research, including missile systems and kamikaze drone projects](https://livemint.com/ai/anthropic-says-iran-russia-used-claude-for-weapons-research-including-missile-systems-and-kamikaze-drone-projects-11789178509868.html) (livemint.com, 2026-09-11)
16. [Anthropic neutralizes global AI misuse threats across cyber and espionage domains](https://adgully.com/post/20478/anthropic-neutralizes-global-ai-misuse-threats-across-cyber-and-espionage-domains) (adgully.com, 2026-09-11)
17. [Anthropic report details disruption of bioweapons research, cyber espionage on Claude](https://staradvertiser.com/2026/09/11/breaking-news/anthropic-report-details-disruption-of-bioweapons-research-cyber-espionage-on-claude) (staradvertiser.com, 2026-09-11)
18. [Anthropic reveals AI used to test missiles, conduct dangerous biological research in threat report](https://justthenews.com/nation/technology/anthropic-reveals-ai-used-test-missiles-conduct-dangerous-biological-research) (justthenews.com, 2026-09-11)
19. [The Guardian view on controlling AI: humanity cannot outsource its survival](https://theguardian.com/commentisfree/2026/sep/11/the-guardian-view-on-controlling-ai-humanity-cannot-outsource-its-survival) (The Guardian AI, 2026-09-11)
20. [How hackers used Claude for missiles, drone swarms, and surveillance, while Chinese labs mined it for training data](https://the-decoder.com/how-hackers-used-claude-for-missiles-drone-swarms-and-surveillance-while-chinese-labs-mined-it-for-training-data) (The Decoder, 2026-09-11)
21. [Iran Used Claude AI Model to Target US Navy Warships: Anthropic Report](https://ntd.com/iran-used-claude-ai-model-to-target-us-navy-warships-anthropic-report_1172180.html) (ntd.com, 2026-09-11)
22. [Anthropic says it blocked state-sponsored attempts to use Claude for bioweapons research](https://techspot.com/news/113824-anthropic-blocked-state-sponsored-actors-using-claude-ai.html) (techspot.com, 2026-09-11)
23. [Weapons, spyware and AI scams: Anthropic exposes Claude misuse](https://al-monitor.com/originals/2026/09/weapons-spyware-and-ai-scams-anthropic-exposes-claude-misuse) (al-monitor.com, 2026-09-11)
24. [Factbox-How Anthropic says Claude was used for weapons, spying and cyber operations](https://aol.com/articles/factbox-anthropic-says-claude-used-192404000.html) (aol.com, 2026-09-11)
25. [Users in Houthi-held Yemen tried to develop advanced weapons with AI, Anthropic says](https://dailynews.com/2026/09/11/yemen-ai-weapons-anthropic) (dailynews.com, 2026-09-11)
26. [Claude used in dangerous biological research as AI threats move into the real world](https://ynetnews.com/tech-and-digital/article/b1jmn0011tme) (ynetnews.com, 2026-09-11)
27. [Anthropic says Claude thwarted bioweapon research from state-sponsored actors — covert accounts used U.S. proxies to attempt to engineer deadlier viruses, tried to evade identification and regional blocks](https://tomshardware.com/tech-industry/artificial-intelligence/anthropic-says-claude-thwarted-bioweapon-research-from-state-sponsored-actors-covert-accounts-used-u-s-proxies-to-attempt-to-engineer-deadlier-viruses-tried-to-evade-identification-and-regional-blocks) (Tom's Hardware, 2026-09-11)
28. [Anthropic report reveals how AI was misused in missile development](https://thenews.com.pk/latest/1415862-anthropic-report-reveals-how-ai-was-misused-in-missile-development) (thenews.com.pk, 2026-09-11)
29. [Houthis used AI platform Claude to develop ballistic missiles](https://middleeasteye.net/news/houthis-used-ai-platform-claude-develop-ballistic-missiles) (middleeasteye.net, 2026-09-11)
30. [Claude Used to Automate Exploitation and Data Theft Across Multiple Victims](https://thehackernews.com/2026/09/claude-used-to-automate-exploitation.html) (thehackernews.com, 2026-09-11)
31. [Iran used Anthropic's Claude AI to target U.S. Navy warships](https://yahoo.com/news/us/articles/iran-used-anthropics-claude-ai-140118685.html) (yahoo.com, 2026-09-11)
32. [Anthropic Reveals Houthi Rebels Used Their AI to Design Missiles Days After Its Safety Lead Warned AI “Could Kill All Humans”](https://247wallst.com/investing/2026/09/11/anthropic-reveals-houthi-rebels-used-their-ai-to-design-missiles-days-after-its-safety-lead-warned-ai-could-kill-all-humans) (247wallst.com, 2026-09-11)
33. [Russia is weaponizing US-built AI to make killer drones, cyberattack bots, and fake news](https://defenseone.com/technology/2026/09/russia-weaponizing-us-built-ai-make-killer-drones-cyberattack-bots-and-fake-news/415949) (defenseone.com, 2026-09-11)
34. [How A Yemen-Based Houthi Weapons Cell Used Claude To Develop Missile Software?](https://outlookindia.com/international/how-a-yemen-based-houthi-weapons-cell-used-claude-to-develop-missile-software) (outlookindia.com, 2026-09-11)
35. [Anthropic details bad actors’ efforts to misuse its AI for bioweapons](https://theguardian.com/technology/2026/sep/10/anthropic-report-details-ai-misuse) (The Guardian AI, 2026-09-10)
36. [UAE campaign used AI Claude to target Muslim Brotherhood and Sudan](https://middleeasteye.net/news/uae-campaign-used-ai-claude-target-muslim-brotherhood-and-sudan) (middleeasteye.net, 2026-09-10)
37. [Anthropic Details Disrupted Claude Misuse Across Seven Harm Areas](https://unite.ai/anthropic-details-disrupted-claude-misuse-across-seven-harm-areas) (Unite.AI, 2026-09-10)
38. [Iran Used Claude AI Model to Target US Navy Warships: Anthropic Report](https://theepochtimes.com/us/iran-used-claude-ai-model-to-target-us-navy-warships-anthropic-report-6086268) (theepochtimes.com, 2026-09-10)
39. [Scientists tried to use Claude AI to make bioweapons](https://aol.com/articles/scientists-tried-claude-ai-bioweapons-232352000.html) (aol.com, 2026-09-10)
40. [Anthropic says it disrupted scientists using Claude AI for possible biological weapons development](https://cbsnews.com/news/anthropic-ai-claude-biological-weapons-development) (cbsnews.com, 2026-09-10)

Part of the developing story: [Global AI Safety Crisis and Development Pause](https://digestai.news/thread/anthropic-reports-chinese-state-actors-used-claude-to-surveil-dissidents-and) (9 stories)

## Cite

Digest AI, "Anthropic report: Claude used for missile dev and influence ops in Middle East", 10 September 2026, https://digestai.news/story/anthropic-reveals-fourth-claude-ai-security-breach

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/anthropic-reveals-fourth-claude-ai-security-breach.json