turing-test

A philosophical and practical test proposed by Alan Turing to determine if a machine can exhibit intelligent behavior indistinguishable from a human.

6 chapters across 5 books

Brainchildren: Essays on Designing Minds (1998)Daniel C. Dennett

Full Text

This chapter introduces Daniel Dennett's collection Brainchildren, focusing on the philosophy of mind and the question 'Can machines think?' Dennett revisits Alan Turing's original formulation of the Turing test, emphasizing its rigor and the common misunderstandings that have led to underestimating its difficulty and overestimating current AI capabilities. He highlights the test's purpose as a philosophical conversation-stopper and a practical challenge rather than a scientific tool, urging clearer thinking about the cognitive powers of computers and their social implications.

Galatea 2.2 (1995)Richard Powers

Chapter 2

The chapter explores themes of identity, memory, and the challenge of authentic communication through the lens of a literature professor interacting with his students and a colleague involved in artificial intelligence research. The professor reflects on a student's vivid, lived experience and memories that defy conventional understanding, while also engaging in a dialogue about the complexities of creating a machine capable of passing a Turing Test by convincingly simulating human intelligence. The narrative intertwines personal storytelling with philosophical and technical discussions about intelligence, belief, and the nature of reality.

Great Ideas in Information Theory, Language and Cybernetics (1966)Jagjit Singh

chapter XIII TURING MACHINES VON NEUMANN'S warning against identifying computing machines and neuron networks with the animal nervous system is quite oppor- tune in view of the present controversy concerning whether or not machines can think. Take first the digital computer we described in Chapter IX. As we saw, it merely works out what has already been thought of beforehand by the designer and supplied to it in the form of program instructions. In fact, it obeys these instructions as literally as the unfortunate Casabianca boy who remained on the burning deck because his father had told him to do so. For instance, if in the course of a computation the machine requires the quotient of two numbers of which the divisor happens to be zero, it will go on, Sisyphus-like, trying to divide by zero forever unless expressly for- bidden by prior instruction. A human computer would certainly not go on dividing by zero, whatever else he might do. The incapacity of the machine to deviate one whit from what the "moving finger" of prior instructions has already decreed make? it necessary to think out in advance every possible contingency that might arise in the course of the work and give the machine appropriate instructions for each case. Is the limitation imposed by this conservation law whereby the machine is incapable of originating a new instruction or idea destined to remain forever? Or will designers be able to construct in course of time automata able to do their own thinking like Rossum robots in Karel Capek's R.U.R. or the Great Brains in Olaf Stapledon's First and Last Men? There are no unequivocal answers to these questions. It all depends on the meaning we choose to assign to the verb "to think." If we adopt a behavioristic definition of the term as the English logician A. M. Turing and his followers suggest, we may consider a machine capable of "thinking" provided it can be made to imitate a human 184

This chapter discusses von Neumann's caution against equating computing machines and neuron networks with the animal nervous system, emphasizing the limitations of digital computers which strictly follow pre-programmed instructions without deviation. It explores the debate on whether machines can truly 'think,' highlighting Turing's behavioristic definition of thinking as imitation of human responses, while questioning if this suffices for genuine understanding. The chapter also introduces the concept of Turing machines as idealized computing devices capable of simulating any other automaton, addressing the complexity and reproduction of automata in relation to biological evolution and information theory.

Brainchildren Unknown

Full Text

This chapter serves as an introduction and overview to Daniel C. Dennett's book Brainchildren, focusing on the philosophy of mind and the question of machine intelligence. It discusses the significance and interpretation of the Turing test as a philosophical and practical benchmark for machine thinking, emphasizing its original intent as a conversation-stopper rather than a scientific tool. Dennett critiques common misunderstandings of the test and highlights the importance of clear thinking about the cognitive capacities of computers, especially as they become integrated into sensitive social roles.

The Age of AI: And Our Human Future (2021)Henry A. Kissinger, Eric Schmidt, Daniel Huttenlocher

Chapter 3

Chapter 3 traces the evolution of artificial intelligence from its origins with the Turing test and early rule-based systems to the modern era of machine learning and neural networks. It highlights the shift from encoding explicit human knowledge to enabling machines to learn from data, overcoming prior limitations in tasks like visual recognition and language translation. The chapter also discusses the revolutionary qualities of AI such as imprecision, dynamism, emergence, and learning capacity, illustrating these with examples like AlphaZero and AI-driven drug discovery.