Misuse of AI
REPORTER:
AI giant Anthropic says it has stopped bad actors from using its AI models to help with cyber threats and biological weapon development.
A report from the firm said actors linked to China, Russia and Iran, as well as criminal groups, were attempting to use the AI model Claude for cyberattacks, bioweapons, surveillance and propaganda. The report also noted that the technology can allow less-skilled people to execute more sophisticated attacks. The announcement comes amid warnings from industry insiders that AI companies must improve ethical oversight.
---
Misuse of AI
REPORTER:
US AI giant Anthropic says its Claude models had been used for malicious purposes.
In a report released Thursday, the company said actors linked to China, Russia, Iran and other countries, as well as criminal groups, bypassed access restrictions and used Claude for potentially dangerous activities, including cyberattacks, weapons development, surveillance and propaganda.
Anthropic also noted AI can now carry out those tasks more independently, making it easier for less-skilled people to perform sophisticated attacks.
The report covers activity the company detected and stopped between December 2025 and August 2026.
00:35
03:25
00:26
03:49
00:32
00:30
00:46
00:35
02:44
02:28
00:34
02:54
00:56
00:43
00:36
02:46