Artificial and Biological Intelligence: Hardware vs Wetware

Abstract

By starting from physical principles, the paper formulates the basis of the biological intelligence and coscience formation. The work shows that far from equilibrium, the principle of maximum energy dissipation, in a fluid phase, brings the molecules to organize themselves in living systems (ordered stationary states) that sustain energy-matter fluxes that determine the architecture of the dissipative system. The driver of the process is the energy; the matter is the substrate. On this base, and thanks to a “selection rule”, the living systems developed a bi-phasic structure consisting of a solid network permeated by a liquid where fluxes of ions and molecules can be maintained. In this way they realize a “wetware” where the energy and the organization/information are handled, and where a high specialization of functions can be obtained by using the spatial delocalization leading to the development of the complexity of the shape. The outputs of a living system can be divided in two categories: 1) macroscopic outputs: Movements, definition of an organized reaction or plan of action, and 2) microscopic ones: Plastic rearrangement of structures, migration of chemical species and electrochemical dynamics (transportation of matter, rearrangement, sensing, storage and information handling). The biological intelligence is a product of these basic processes where the wetware (matter substrate) is continuously modified by the energy-matter fluxes. On this base, the artificial intelligence of computers can be seen as a subclass of intelligent processes since it owns some limitations due to the invariance of the material substrate (that is not self-modified by the energy fluxes) and by the unique form of the flux of matter that is given by the electron current. Finally, the definition of a bio-inspired artificial intelligence is discussed.