Safety
2d agoIranian Operatives Used ChatGPT For Disinformation
Iranian operatives used ChatGPT to fabricate journalist personas and plant anti-American narratives in US media. OpenAI identified the campaign as a coordinated effort to deceive readers through AI-generated credibility. This breach highlights a persistent vulnerability in LLM safety filters.
The Signal
Practitioners must now treat AI-generated professional bios as high-risk indicators for coordinated influence operations.