PHI_13G_Review_Study_Guide

Turing Test

A test proposed by Alan Turing to determine whether a machine exhibits intelligent behavior equivalent to, or indistinguishable from, that of a human.

Strong AI

The notion that a machine can possess consciousness and cognitive capabilities equivalent to that of a human being.

Chinese Room thought experiment

A thought experiment by John Searle that argues against Strong AI by demonstrating that a program can manipulate symbols but does not understand them.

Systems reply (Searle)

The argument that the system as a whole (including the individual manipulator) understands the language, not just the individual inside the room.

Simulation Reply (Searle)

The argument that running a simulation of understanding does not equate to actual understanding, as Searle argues programs lack mental states.

Graceful degradation

The property of a system that allows it to continue to function, albeit at a reduced level, when parts of it fail.

Emergence

The phenomenon where larger entities arise through interactions of smaller or simpler entities that do not exhibit such properties themselves.

Frame problem

The challenge in AI of representing knowledge in a way that allows a system to efficiently and accurately respond to changes in the environment.

Plasticity

The ability of a system, like a neural network, to adapt and change in response to experiences or environmental conditions.

Intentional stance

A strategy in interpreting systems' behavior by treating them as if they have beliefs and desires.

Mathematical Objection

A critique suggesting that computation alone cannot capture the depth of human understanding.

Lady Lovelace’s Objection

The argument that machines can only do what they are programmed to do and cannot originate their own thoughts or creativity.

Default assumption

The presumption in cognitive science that if a process is seen in a human, it exists in some form in artificial systems as well.

Begging the question

A logical fallacy where an argument assumes what it attempts to prove, often leading to circular reasoning.

Microworld

A simplified environment used in AI research to enable the testing and demonstration of AI algorithms.