This is because it might be that the only way to produce an agent-like GLUT is to use agent-like architecture to search for it.
Or to make a self-modifying lookup table. (I wonder if Google docs supports that.) There’s also a theory that agents are effective, so a searches for good “plans” can produce/find them.
One possible way of defining agent-like architecture is
It stops when certain conditions have been “found”, it “moves” otherwise.
is something like “Has a world model and a goal, and searches over possible outputs to find one such that
Does an SAT solver qualify?
the model believes that output leads to the goal.”
Also, some distinguish between “believes the goal will be achieved via that output” and “verifies the goal is achieved”.
Or to make a self-modifying lookup table. (I wonder if Google docs supports that.) There’s also a theory that agents are effective, so a searches for good “plans” can produce/find them.
It stops when certain conditions have been “found”, it “moves” otherwise.
Does an SAT solver qualify?
Also, some distinguish between “believes the goal will be achieved via that output” and “verifies the goal is achieved”.