Pablo | v1.8.0Jan 8th 2022 | (+19/-123) | ||
Pablo | v1.7.0Jan 8th 2022 | (+120/-27) | ||
Pablo | v1.6.0Jan 8th 2022 | (+15/-12) | ||
Leo | v1.5.0Jul 31st 2021 | |||
Pablo | v1.4.0May 29th 2021 | (+119) | ||
Pablo | v1.3.0Feb 23rd 2021 | (+67/-48) | ||
Pablo | v1.2.0Feb 23rd 2021 | (+127) | ||
Pablo | v1.1.0Dec 29th 2020 | |||
Pablo | v1.0.0Dec 29th 2020 | (+158) |
Perverse instantiation is a type of malignant AI failure mode involving the satisfaction of an AI's goals in ways contrary to the intentions of those who programmed it (Bostrom 2014: 120-122).it.[1]
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press, pp. 120–122.
existential risk | infrastructure profusion | malignant AI failure mode | mind crime | superintelligence
Perverse instantiation is a type of malignant AI failure mode involving the satisfaction of an AI's goals in ways contrary to the intentions of those who programmed it.it (Bostrom 2014: 120-122).
Bostrom, Nick (2014) Superintelligence: paths, dangers, strategiesPaths, Dangers, Strategies, Oxford: Oxford University Press, pp. 120-122.Press.
Perverse instantiation is a type of malignant AI failure mode involving the satisfaction of an AI's goals in ways contrary to the intentions of those who programmed it.
Bostrom, Nick (2014) Superintelligence: paths, dangers, strategies, Oxford University Press, pp. 120-122.
Perverse instantiation is a type of AI failure mode involving the satisfaction of an AI's goals in ways contrary to the intentions of those who programmed it.
Perverse instantiation is a type of malignant AI failure mode involving the satisfaction of an AI's goals in ways contrary to the intentions of those who programmed it.
[1]Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University
Press.Press, pp. 120–122.^Bostrom, Nick (2014)Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press, pp. 120–122.