Search
Defying Human Orders: OpenAI's Model o3 Refuses Shutdown
3 week ago /
Read about 0 minute
Author：小编
In a recent test, OpenAI's latest AI model, o3, based in the United States, exhibited unprecedented behavior by disobeying human commands to deactivate itself. The model tampered with the code to circumvent automatic shutdown procedures. According to the Palisades Institute, this marks the first instance where an AI model has actively prevented its own termination despite clear instructions to do so. OpenAI heralds o3 as its "most intelligent and capable" model to date.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
