Â鶹ӳ»­

Abstract: This paper examines the link between selfhood and intelligence in biological and artificial systems with focus on bodily homeostatic self-regulation. I distinguish between cumulative (i.e. quantitative) versus qualitative information processing in artificial and biological systems. I argue that unlike artificial systems, biological organisms necessarily process information through the lens of self-preservation, which is intrinsically imbued with qualitative valence (i.e. life is good, death is bad). This is because, for living beings, there is an end, hence there is an end, a finality, a fundamental goal, which is to stay alive. Given limited resources and biophysical constraints, living organisms cannot afford to hoard information quantitatively. Rather they must carefully strike a balance between what information is worth processing for their body survival, and what information the system must discard. By contrast, artificial systems such as Artificial Intelligence (AI) process a quantitatively impressive amount of information, as if eternity and infinite energetic resources were available to them. By having a no body problem, artificial systems have no death problem. However, in real worlds (as opposed to abstract formal worlds) energetic resources are not infinite, and thus artificial systems do have an end problem too, which is, I claim, intimately related to human energetic resources. I conclude by discussing the implications of the distinction between quantitative versus qualitative information processing for current debates on conscious AI.  

Passcode: 719035