SEARLES CHINESE ROOM ARGUMENT
John Searles Target: Strong AI
Even if we grant for the sake of argument 2 common assumptions of strong AI:
1. Linguistic output indistinguishable from a human.
2. Turing test passed.
A computer still cannot understand anyt
THE CASE FOR ARTIFICIAL INTELLIGENCE
The Dartmouth Conference (The Dartmouth Summer Research Project on AI)
Claude Shannon (AT&T)
John McCarthy (Dartmouth) coins Artifical Intelligence
Marvin Minsky (MIT)
Nathaniel Rochester (IBM)
Allen Newell and Her
AUTOMATA THEORY AND TURING MACHINES
Transition Diagram: States contain transitions. Transition conditions.
Definition of Automata Theory
Body of physical and logical principles
Underlying the operation of any actual or virtual computational device (an
THE CONSCIOUSNESS OBJECTION
Recall Daniel Dennetts Three Stances
Physical Stance (physics & chemistry)
Design Stance (biology & engineering)
Intentional Stance (software & minds)
o But does this require consciousness?
Three Senses of Consciousness
Basic Philosophical Areas
Value Theory (Axiology)
History of Computers
DNA: programmed function but not technology
Abacus: technology but not automatic
Quipo: technology but not automatic
Francisco Pizzaro and
Models for Computing
1. Classic AI GOFAI
2. Connectionism Artificial Neural Networks (ANN) aka Parallel Distributed
3. Robotic architecture
Learns through feedback
o Nodes vary in firing rate
o Input, hidden, and ou
THE SIMULATION ARGUMENT
At least one of the following propositions is true:
o The human species is very likely to go extinct before reaching a posthuman
o Any posthuman civilization is extremely unlikely to run a significant number of
THE TURING TEST
Can a machine think? (Ontological issue)
How would we know? (Epistemological issue)
Grandfather of AI?
For REASON, in this sense, is nothing but reckoning (that is, adding and
subtracting) of the consequences of general nam
THE CREATIVITY OBJECTION
1. Real minds have creativity.
2. Computers do not and never will have creativity.
Therefore, computers can never have real minds and strong AI is not possible.
1992 American Association of AI conference
THE FREE WILL OBJECTION
The Prediction Version of the Free Will Argument
1. Computers only do what they are programmed to do.
2. Programming is done by human programmers who know (can predict) what their
programs will do.
3. Anything that can be predicted