Disrupting malicious uses of AI
OpenAI
This report details case studies on detecting and preventing the malicious use of AI by threat actors.Summary
This report presents case studies illustrating how malicious actors are leveraging AI models, often in conjunction with traditional methods like websites and social media. Over the past two years, analysis has revealed that threat activity rarely confines itself to a single platform or AI model; instead, actors utilize various AI tools throughout their operations, as demonstrated by a case study involving a Chinese influence operator. The purpose of sharing these insights is to enhance the industry's and society's ability to identify and mitigate these evolving threats.
(Source：OpenAI)