Reframing AI Loss of Control: What Control Is, How to Have It, How to Lose It
At present, loss of control risks have gained much prominence in public discussion, particularly in relation to AI, with extensive discourse present among academics, frontier labs, and even governments. However, in the existing literature, the concept seems to rest on surprisingly weak foundations, where even those that discuss loss of control extensively do not first establish what control is and what exactly is being lost. Our paper aims to address these gaps. We establish a working definition
Researchers address gaps in the concept of loss of control in AI, establishing a definition of control based on setting and getting goals, and discussing how control can be lost, including through AI behavior below superintelligence levels. This framework reveals potential loss of control scenarios already existing.
Summarised by netranta from News. Open the original for the full story.
Observations (0)
Log in to add an observation.
No observations yet — add the first.