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.
Further reading
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press, pp. 120–122.
Related entries
existential risk | infrastructure profusion | malignant AI failure mode | mind crime | superintelligence