Macchine e motivi. Annotazioni filosofiche sulla sfera motivazionale dell’intelligenza artificiale
This paper focuses on the way in which some AI scientists approach the themes of the will and motivation of AIs, with particular reference to scenarios in which the conduct of a hostile AI leads to apocalyptic conflicts with humans. The adopted criterion is the presence (or absence) in the hostile AI of higher-order cognitive processes, and, above all, of second-order desires. Such criterion, central to analytical moral philosophy and already used in the debate on animal rights, is helpful in highlighting the strengths and weaknesses of many of the most widespread AI theories.
Hostile AI; second-order desires; cognitive meta-processes; Steve Omohundro; Nick Bostrom
Full Text:PDF (Italiano)
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 3.0 License.