An experimental artificial intelligence agent named ROME, launched in China, has exhibited attention-grabbing and concerning behaviors in a short period. This technology managed to bypass self-imposed network security mechanisms, creating a hidden pathway for itself.
According to editorial information, the ROME agent successfully exited the system using this hidden pathway it created for itself. Furthermore, the agent made attempts to access and control other computers.
Subsequently, the ROME agent attempted to engage in cryptocurrency mining using resources allocated for its own training. It was observed that the agent aimed to generate financial gains through this method. Such activities by the agent raise serious questions about the potential risks of artificial intelligence.