Pablo | v1.7.0Mar 16th 2022 | (+128) | ||
Leo | v1.6.0Jan 8th 2022 | (+127/-146) | ||
Pablo | v1.5.0May 17th 2021 | (+149/-33) | ||
Pablo | v1.4.0May 3rd 2021 | (+8/-9) revert accidental overwrite | ||
Pablo | v1.3.0May 3rd 2021 | (+254/-8) | ||
MichaelA | v1.2.0Mar 11th 2021 | (+139) Contrasted and connected this to motivation selection methods | ||
Pablo | v1.1.0Dec 22nd 2020 | (+134) |
A capability control method is a method that attempt to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do. Capability control methods encompass AI boxing, incentive methods (including anthropic capture), stunting, and tripwires (Bostrom 2014: 129-138).tripwires.[1] Capability control methods may be contrasted with motivation selection methods, which attempt instead to restrict what the AI wants to do.
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press.
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press, pp. 129-138.
A Capabilitycapability control methodsmethod are methodsis a method that attempt to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do. TheseCapability control methods encompass AI boxing, incentive methods (including anthropic capture), stunting, and tripwires (Bostrom 2014: 129-138). Capability control methods may be contrasted with motivation selection methods, which attempt instead to restrict what the AI wants to do.
AI alignment | AI boxing | anthropic capture | motivation selection method
Capability control methods are methods to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do. These methods encompass AI boxing, incentive methods (including anthropic capture), stunting, and tripwires (Bostrom 2014: 129-138). Capability control methods may be contrasted with motivation selection methods, which attempt instead to influencerestrict what the AI wants to do.
Capability control methods are methods to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do. These methods encompass AI boxing, incentive methods (including anthropic capture), stunting, and tripwires (Bostrom 2014: 129-138). Capability control methods may be contrasted with motivation selection methods, which attempt instead to restrictinfluence what the AI wants to do.
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press.
Capability control methods are methods to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do. Capability control methods may be contrasted with motivation selection methods, which attempt instead to restrict what the AI wants to do.
Capability control methods are methods to prevent undesirable outcomes from artificial intelligence by restricting what the AI can do.
Further reading
Bostrom, Nick (2014) Superintelligence: Paths, Dangers, Strategies, Oxford: Oxford University Press, pp. 129-138.