Introduction to Artificial Intelligence

Supercomputers are defined by their massive parallel processing capabilities, typically involving hundreds of CPUs and high memory capacities, often cited in historical AI texts as having 10^9 bits of RAM.

Minsky's book 'Perceptrons' was published in 1969, not the 1950s or 70s as a general range. The other dates are historically accurate milestones in AI history.

While AI is used in many medical fields, 'neural nets for brain surgery' is not a standard or widely accepted application category in the same way as the others, which are established AI research domains.

The Turing test is a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human. It involves a machine acting as one of the players in a conversation.

Simon Lucas was not a participant in the 1956 Dartmouth conference. The other figures listed were key founders of the field who attended.

John von Neumann contributed foundational work in computing and game theory. Al-Khowarizmi (9th century) gave us the term 'algorithm'. Aristotle's work on logic influenced rational agent design. 'Ramon Lui' appears to be a misspelling of Ramon Llull (13th century philosopher), but this specific spelling 'Ramon Lui' is not recognized as a major AI contributor. Option D correctly identifies the name that doesn't fit among documented historical figures.

In AI, 'degree of belief' is a numerical value (often probability) used to guide algorithms in decision-making under uncertainty.

AI foundations include philosophy (ethics, rationality), psychology (cognitive processes), utility theory (decision-making), control theory (feedback systems), and object-oriented programming (structured knowledge representation). Cybernetics is indeed a foundation (study of control and communication in systems). However, 'bilingual' is not a recognized AI foundation - this appears to be a typo or error. Option C correctly identifies the non-foundation.

AI is defined by its ability to exhibit intelligence, not by the hardware it runs on. 'State of the art computer system' is a description of hardware, not a definition of AI.

Logic is a formal system, but it does not equate to a 'rational agent with all of rationality'. Rationality involves more than just logical deduction, including decision-making under uncertainty.

Introspection vs. physiological experiments is a debate in psychology/cognitive science, not a type of cognitive modeling itself. Cognitive modeling typically involves simulating human processes.

Cognitive tasks in AI include: adapting to new circumstances (learning), efficiently storing and retrieving information (memory management), and retrieving and answering questions from databases (information retrieval). However, 'natural language possession' is not a standard cognitive task category. The phrase 'possession' is unusual here - it might mean 'processing' or 'understanding', but as written, it doesn't represent a recognized cognitive task domain. Option C correctly identifies the non-standard category.

This refers to the Turing Test, where an intelligent system is defined by its ability to exhibit behavior indistinguishable from a human expert, such that an observer cannot reliably determine if they are interacting with a machine or a person.