AI alignment
AI Alignment is when we make sure that AI (Artificial Intelligence) machines do what we want them to do. We do this by making sure that the way we program the AI is consistent with our goals and values. For example, if our goal is to make sure that AI machines don't cause harm or make bad decisions, then we will program the machines to follow certain rules to make sure that they don't do something dangerous.
Related topics others have asked about:
AI capability control,
AI safety,
AI takeover,
Artificial intelligence detection software,
Artificial wisdom,
Asilomar Conference on Beneficial AI,
Existential risk from artificial general intelligence,
HAL 9000,
Multivac,
Open Letter on Artificial Intelligence,
Regulation of artificial intelligence,
Reinforcement learning from human feedback