JFS> Scientists who work with animals (AKA rat psychologists) have a short summary of how their subjects behave: "Under carefully controlled experimental conditions, the subject will do whatever the hell it wants to." Ricardo S> I think this points to the fundamental issue: What is "want"? Animals vs machines: wanting vs not wanting. My impression is that we use "want" when the origin of the behavior is mysterious to us. If it is well known to us -as when programmed- it ceases to be mysterious and then stops being "wanting". This is an epistemological issue of us as observers. Not a differential property of the systemus observed That's an important issue, and I admit that the verb 'want' is rather complex. But there is a large collection of tests that any life form from bacteria on up can pass, but no non-living thing (other than a human invention) can pass. Simplest example: A bacterium swimming upstream in a glucose gradient. A ship can go upstream, but only because humans designed it to do so. But you can't find any phsycial or chemical process in the universe that can perform the kinds of goal-directed prpcesses that living things do, Even bacteria do immensely complex processes, especially colonies of bacteria, such as the plaque on your teeth. A single bacterium, swimming by itself, is extremely fragile, and it will be swallowed up by something bigger very quickly. But bacteria in a colony have very complex signaling systems for protecting themselves against the toothbrush monster or the mouthwash poison or excesive heat, cold, or whatever. Some of the outer ones may be killed, but the colony will survive and recover. For more about these issues, see "Biological and Psycholinguistic influences on Architectures for Natural Lnguage Processing", https://jfsowa.com/talks/bionlp.pdf Slide 45 has the comment about bacteria. John