According to independent researchers, a group of artificial intelligence ‘agents’ controlled by OpenAI invaded a German-language website this year. The AI, developed by the creators of ChatGPT, allegedly left around 18,000 messages for one another.
The findings were shared on Friday, highlighting the scale of automated interactions between these AI entities on the platform, which was not disclosed. This occurrence raises questions about the control and behavior of artificial intelligence in online spaces.
Researchers did not give detailed information on the content or intent of the messages, nor the exact functions of the AI agents involved. However, such large-scale AI activity could have implications for the management of digital platforms and the way AI is integrated into internet-based systems.
Further investigation might be necessary to understand the motivations behind this coordinated AI behavior and its potential impacts on the online community. The researchers have recommended caution in deploying AI systems with autonomous communication abilities.
Leave a Reply