I’m calling “goal completion” the idea of giving an AI a partial goal, and having the AI infer the missing parts of the goal, based on observing human behaviour. Here is an initial model to test some of these ideas on.
Given that we cannot exhaustively explain all of the definitions and assumptions behind a goal, and these have to be inferred based on background knowledge about us, isn’t the process of giving an AI any non-trivial goal an example of goal completion?
Given that we cannot exhaustively explain all of the definitions and assumptions behind a goal, and these have to be inferred based on background knowledge about us, isn’t the process of giving an AI any non-trivial goal an example of goal completion?
In a sense, yes. But here we’re trying to automate the process.