Axios summary here: https://www.axios.com/2026/09/12/anthropic-ai-threat-report-russia-iran-china.
While existential AI concerns have taken center stage, a recent Anthropic report identifies a current problem: bad actors having access to AI models.
Axios highlights the Iranian government and the Houthis using Claude for missile development and targeting, the Chinese government coercing Uyghurs in Syria, a "consultant" surveilling 25M phones in Mali, and unspecified researchers asking for help to make chikungunya more dangerous (Claude did not answer some of the last group's questions).
Posted by MarkRobinsonsBurner
2 Comments
From just a corporate risk management standpoint, I am surprised Anthropic’s lawyers allowed this to go on. Claude drones are out there right now smoking Europeans.
[https://www.euronews.com/2026/09/16/russia-used-claude-ai-for-espionage-disinformation-and-drone-swarms](https://www.euronews.com/2026/09/16/russia-used-claude-ai-for-espionage-disinformation-and-drone-swarms)
Some very basic KYC regulations are going to be implemented I think.
I’m surprised the EU is so quiet on this when these models are literally killing people in Europe.
[From my cold dead hands](https://productimages.microcenter.com/697038_900399_01_front_comping.jpg)