Computing machinery and intelligence

by Alan Turing

"Computing Machinery and Intelligence" is a seminal paper written by Alan Turing on the topic of artificial intelligence. The paper, published in 1950 in Mind, was the first to introduce his concept of what is now known as the Turing test to the general public.

Published in Mind 49: page 433-460.

(Source: Wikipedia)

Genres: Science, Philosophy, Nonfiction, Computer Science, Artificial Intelligence, Technology, Programming

24 pages, Unbound

First published January 1, 1950

About the author

Alan Mathison Turing (23 June 1912 – 7 June 1954), was an English mathematician, logician, cryptanalyst and computer scientist. He was highly influential in the development of computer science, providing a formalisation of the concepts of "algorithm" and "computation" with the Turing machine, which played a significant role in the creation of the modern computer. Turing is widely considered to be the father of computer science and artificial intelligence.

During the Second World War, Turing worked for the Government Code and Cypher School at Bletchley Park, Britain's codebreaking centre. For a time he was head of Hut 8, the section responsible for German naval cryptanalysis. He devised a number of techniques for breaking German ciphers, including the method of the bombe, an electromechanical machine that could find settings for the Enigma machine. After the war he worked at the National Physical Laboratory, where he created one of the first designs for a stored-program computer, the ACE.

Towards the end of his life Turing became interested in mathematical biology. He wrote a paper on the chemical basis of morphogenesis,[3] and he predicted oscillating chemical reactions such as the Belousov–Zhabotinsky reaction, which were first observed in the 1960s.

Turing's homosexuality resulted in a criminal prosecution in 1952, when homosexual acts were still illegal in the United Kingdom. He accepted treatment with female hormones (chemical castration) as an alternative to prison. He died in 1954, several weeks before his 42nd birthday, from cyanide poisoning. An inquest determined it was suicide; his mother and some others believed his death was accidental. On 10 September 2009, following an Internet campaign, British Prime Minister Gordon Brown made an official public apology on behalf of the British government for the way in which Turing was treated after the war.

Reviews

[Javier Maldonado · Rating 5 out of 5]

Un texto fundacional, con un valor histórico tremendo, donde Turing aborda temas como la computación y la programación (aún en pañales en aquella época), la inteligencia artificial e incluso la filosofía de la mente. Creo que este ensayo (publicado por Turing en la revista "Mind" en 1950) necesita ser estudiado y no solo leído, sobre todo por la complejidad de algunos conceptos e ideas y porque no está escrito para un lector casual. En cuanto a la presente edición, encuentro brillante y agradezco la decisión de haber incluído no solo la traduccción sino que también el texto en su idioma original (inglés).

[Thomas Herzog · Rating 5 out of 5]

In this classic essay, Turing makes the case for the possibility of thinking machines. He does this mainly by listing possible objections to this claim, and showing why none of them are, on closer inspection, very convincing. (Fun facts: He has little patience for theological arguments, yet is strangely worried how ESP phenomena, specifically telepathy, might be taken into account.)

His main proposal for deciding if a machine can think is by a kind of game, "The Imitation Game", what is now commonly called the "Turing Test": A (human) interrogator, after interacting for a given time only via text (i.e. through a keyboard and screen) with two other unseen participants, is asked which of them is human. Turing even makes a quite specific prediction:

I believe that in about fifty years' time it will be possible, to programme computers, with a storage capacity of about 10^9, to make them play the imitation game so well that an average interrogator will not have more than 70 per cent chance of making the right identification after five minutes of questioning.

While this might have been overly optimistic in terms of time frame and storage requirements, in general the essay is highly prescient. Turing anticipates the importance of machine learning, and everyone familiar with current developments around "deep learning" will recognize this phenomenon:

An important feature of a learning machine is that its teacher will often be very largely ignorant of quite what is going on inside, although he may still be able to some extent to predict his pupil's behavior.

All in all, a very stimulating read, and highly recommended.

[Margareta · Rating 5 out of 5]

It's marvelous how the human mind could predict such a great helper as the AI: from 1843, the first time by Ada Lovelace and Charles Babbage to 1950 by Alan Turing and to modern days, where humans can't even compete anymore.
The description of the problem "Can machines think?" in this paper is a very thorough one and tackles both positive and negative sides of the problem, proving that in the near future - the '00s, humans would have achieved the needed storage space and developed technology to such a level that it would be possible for a machine to learn chess or English. It's impressive how well Alan Turing predicted all that, especially the timing and the fields in which AI would be used in the first places. And he was right - in '96, the first AI beat Gary Kasparov at chess. Today, we use things like Alexa, Siri, and Cortana to make sense of our vocal and text messages to get us the information that we need and to do simple tasks like a to-do list or message someone.
I can only wonder to what extent will AI get proficient and what it could create in the future, given that it already can generate songs and abstract paintings.

[Victoria Oldemburgo De Mello · Rating 4 out of 5]

Alan Turing: the argument that machines can’t think because they can’t surprise people is weak because they surprise me all the time.
Also Alan Turing: yeah you got me… machines can’t be telepathic so maybe they can’t REALLY think… but what if you're in a telepathy-proof room?

[L-ssar · Rating 5 out of 5]

En este corto ensayo se puede apreciar la genalidad de Turing: deja el germen de la IA y redes neuronales, nos habla de matemáticas de sistemas caóticos e incluso de telepatía (que el creía posible como fenómeno aun no explicado).

[Raoul G · Rating 5 out of 5]

This seminal 1950 work by Alan Turing is a great pleasure to read. In it he approaches the question "Can machines think?". Today, in the age of machine learning and artificial intelligence, this questions definitely seems warranted, but to see that somebody approached this question in a way that is still relevant more than 70 (!) years ago, is truly fascinating. This goes to show how ahead of his time Turing really was.

How does Turing approach this question? He approaches it by presenting what he calls the "Imitation Game". This thought experiment, in one form or another, is quite well-known nowadays, as it was popularized by different movies and books. Simplified it can be stated as following: Can a machine imitate human intelligence well enough to convince a human interrogator that it is human? As a rigorous scientist, Turing defines all parts of the thought experiment well and responds to possible objections.

In this work one can also find a short explanation of the digital computer and how it relates to the human brain. Furthermore, discrete state machines are mentioned and their relation to the question at hand is illuminated. While not giving a final response to the main question of this work, the following quote shows where Turing thought things were heading:
"...I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted."

Turing believed that reaching the state where we are able to produce machines that "think" is not problem of engineering, but rather of programming. He also proposed an approach to how the machine could be programmed in order to be able to beat the imitation game. What astounds me is how many of the concepts he developed here are the basis for techniques used in artificial intelligence today (e.g. machine learning, evolutionary algorithms, reinforcement learning). He also predicted that we will at some point not be able to understand how the "learning machine" works internally anymore, which is already the case for deep learning algorithms such as neural networks.

Turing can thus really be seen as a hero of computer science and I recommend this ground-laying work to anyone interested in the domain.

[Basel · Rating 2 out of 5]

Imagine we construct a Turing machine of a man who cannot read where he has a table of colors juxtaposed with wave frequencies of these colors(data base),and when a question handed to him say the word red he draws the numbers of the frequency of the red color(when Red is seen match its shape then print the juxtaposed cell,same to the reverse)(control) and him the can’t read man is the executive unit.we can complicate it a bit and add other information which can juxtaposed with the color and we can keep that going until we can cover the whole available knowledge of colors known to a top tier physicist.the interrogator outside the two rooms cannot tell who is the machine,therefore the man who cannot read ‘understands the physics of colors’ understands it just as our top tier scientist in the field.
If this doesn’t sound a very defective model of understanding and the mind then your mental states and the content you give to symbols and the very transparent difference between the physicist conscious understanding of the physics of colors and the mere computational content free symbols of the man who cannot read are one and the same which least to say is crazy.

[Stacey · Rating 5 out of 5]

I was a little skeptical about the relevance of this article at first, but it took me by surprise. A glimpse into digital computers from the past, refutation of several points of view claiming that the Turing test is not enough. There is even description of the old debugging process and a reference to Lady Lovelace!

Also, being able to predict learning machines in 1950? Artificial Intelligence must have been born the same second he thought about it. And look how much it's grown 71 years later...

[Lidia · Rating 5 out of 5]

lectura obligatoria de la asignatura "filosofía de la inteligencia artificial" (me ha gustado mucho)

[Jose Gaona · Rating 5 out of 5]

"Personalmente creo que, dentro de unos cincuenta años, se podrá perfectamente programar computadoras con una capacidad de almacenamiento aproximada de 1.000.000.000 para hacerlas jugar tan bien al juego de imitación que un preguntador corriente no dispondrá de más del 70 por ciento de las posibilidades para efectuar una identificación correcta a los cinco minutos de plantear las preguntas."

Texto fundacional de la corriente dura en funcionalismo en Inteligencia Artificial y Ciencia Cognitiva. En este texto el bueno de Alan muestra ser, como profeta y anunciador de su religión, menos papista que sus acólitos, a pesar de la cita de más arriba. Turing traduce la pregunta que pone título al libro por la pregunta de si una máquina superará el famoso test que lleva su nombre, para lo cual, pasa a desgranar ese "juego de imitación". Tras analizar las nociones de computador digital y su universalidad y responder a distintas objeciones a su planteamiento, afirma:

"Habrá comprobado el lector que no dispongo de argumento positivo alguno lo bastante convincente para apoyar mi tesis. Si lo tuviera, no me habría tomado tanta molestia en exponer detalladamente las falacias de las tesis contrarias."

En otras palabras: Turing da muestras de honestidad y humildad intelectual y acepta que su planteamiento no es irrebatible, aunque como programa de investigación empírico, tampoco sea refutable conceptualmente. ¿El resultado? Parece ya un lugar común en la literatura especializada que el programa iniciado por Turing hace más de 60 años, de dar como resultado algo, no será la replicación de la inteligencia humana, sino algo más bien diferente. Pero también parece un lugar común, al mismo tiempo, que habremos aprendido muchas cosas sobre la mente humana cuando terminemos ese sendero. Por simple y llana comparación, claro.

[Adil Master · Rating 5 out of 5]

I should first mention that I am a computing student at sixth form and I am a tech head at heart so this is probably a biased review.

Can Machines think? is the question that Alan Turing poses us in this paper and in this amazing and detailed paper, he throughly explain what his answer is and what the power of machines could be. Some of his explination revolves around The Imitaion Game (not the movie) but an actual game were two people are questioned and an interagor tries to guess who is who? However, the interogator has no idea who is who.

This was just a fascinating paper to read. To see how Alan perfectly articlulates what we would call now Artificial Interlligence and machine learning is excellent. I mean he is Alan Turing so all of his theories are excellent and since being a computing student I just wanted to read the paper and I was just similing because I knew I was reading a paper that was written by someone who had shared such a powerful passion of machines and that is something to admire to. I also found that i was fun to make comparisions from when he writes (1950) to the modern day.

Overall, this was a great read.

[Cwpper David · Rating 4 out of 5]

Definitivamente, la voz de un genio. El texto está tan bien estructurado (Ojalá lo hubiera leído antes de presentar mi tesis de grado), su teoría es tan impresionante, que en un alarde de honestidad científica, el autor empieza a exponer las posibles objeciones a sus propias ideas, dando pie a otro remolino de conjeturas intelectuales aún más interesantes, que van desde la teología hasta la percepción extrasensorial, abordando por el camino temas tan controversiales como telepatía, clarividencia, psicocinesis, en fin. Todo un viaje. Incluso llega a sugerir la construcción de una computadora bebé a la que se le pueda enseñar a través de castigos y recompensas. Una genialidad.

Link de Descarga: http://losmenosprecie.blogspot.com/20...

[Nuno R. · Rating 4 out of 5]

An historical paper, that created "The Imitation Game", and was influential in all the AI research and thinking since it was published in 1950. Very well argued. Elegant writing. A pleasure to read.

[Naísia Xavier · Rating 1 out of 5]

Acho importante começar apontando que, na classificação do Goodreads 1 estrela significa simplesmente "não gostei", enquanto 2 significa "it was ok". Não estou dizendo que o livro é um lixo que não serve para nada. Estou dizendo que não gostei. A resenha explica:

Se, ao invés de "Fisiologia", o doutor Victor Frankenstein tivesse se voltado para a computação, é isso que você teria. Só considerações à base do "posso fazer", e ética nenhuma.

A imagem de ser humano de Turing era problemática. Infelizmente, ele não conseguia argumentar sobre o potencial da máquinas sem, ao mesmo tempo, restringir o que é ser uma pessoa. Me pareceu que no pensamento dele, neste artigo, as duas coisas estão em um nó indesatável, onde para dar corda em um do lado, você precisa tirar do outro, porque ela é bastante limitada. Um dos argumentos dele, por exemplo, é que o cérebro de uma criança seria como uma tábula rasa.

Segue ipsi litteris:

"Presumably the child-brain is something like a note-book as one buys it from the stationers. Rather little mechanism, and lots of blank sheets. (Mechanism and writing are from our point of view almost synonymous.) Our hope is that there is so little mechanism in the child-brain that something like it can be easily programmed".

A bem da honestidade intelectual preciso parar e dizer: esse, provavelmente, é o segundo pior argumento no texto inteiro, e não um dos seus pontos fortes. Mas, mesmo sem querer fazer um espantalho do autor, acho que a consequência educacional coletiva da ideia de alguém em relação às crianças de uma sociedade é sempre algo necessário de ser examinado, porque é bem sintomático a respeito das dimensões de futuro que essa ideia pode abrir ou desautorizar. Turing, na realidade, não estava se ocupando da educação das crianças; ele apenas supunha que, com máquinas --- sendo seu pressuposto crianças como uma tabula rasa---- ele poderia fazer igual.

Mas, definitivamente, o que se encontra aqui não é uma boa concepção de educação, tendo em mente uma sociedade possivelmente mais humanizada... e muito menos o que pai e mãe costumam julgar a partir daquele argumento indefensável, porque inquestionável: amor. Mas fica a questão para ser indagada aos colegas transumanistas...

Muita coisa proposta por ele foi de fato realizada com o advento da Inteligência Artificial, a respeito da qual ele estava certo, e é mais por isso que o texto permanece. Isso e a inestimável importância histórica de seu papel como cientista descriptografador, para vencer os nazistas na Segunda Guerra.

Mas, em que pese sua competência técnica, Turing, escrevendo, emerge como um pensador ainda meio"grosseiro", que aparentemente não parou observar nenhum dilema do tipo fato X valor, em relação às questões de que trata aqui.

Seus argumentos por assim dizer "ontológicos" têm mais a ver com mera lógica formal e/ou uso da linguagem do que com considerações a respeito de uma realidade multifacetada, complexa e processual. A vertente determinista do empirismo e o naturalismo científico --- tradições, infelizmente, bastante britânicas, aliás ---- vêm à mente.

Porém, quanto aos argumentos computacionais, a concepção dele sobre o cérebro/máquina no que diz respeito, especificamente, às suas ideias sobre quantidade de dados e armazenamento de dados, como critério para a possibilidade da "imitação", máquinas com capacidade armazenamento de 10 à 9ª potência ficou datada em viabilização da singularidade, de uma forma que faz com que o argumento precise ser revisto em termos numéricos de viabilidade empírica, como veio a demonstrar esse experimento da Google com Havard, em foi preciso 1,4 petabyte (ordem de 10 à 15ª) para mapear apenas 1 centímetro cúbico do cérebro humano em imagens nanométricas https://olhardigital.com.br/2021/06/0.... Os responsáveis declaram que a fronteira agora é justamente a capacidade armazenamento, que possivelmente só será alcançada com processadores quânticos e/ou nanoarmazenamento. Mapear um cérebro completo iria requerer um exabyte, ou seja, um bilhão de gigas, 10 à 18ª potência. Para um cálculo feito em 1950, acho que é justo dizer que foi um chute admiravelmente certeiro, embora, ainda inexequível após 70 anos da previsão que ele supunha concretizável em meio século.

Também é importante notar que ele parece ser um proto-proponente do que a Filosofia, em seu ramo "Filosofia da Mente" chama de "Teoria da Identidade Mente Cérebro", ou materialismo reducionista/materialismo eliminativo. O fato é que ele simplesmente não dispunha de argumentos a respeito do problema da causação mental, que aliás, permanece em aberto, do ponto de vista científico e quanto à Filosofia da Mente, também, de acordo com autores como André Leclerc, Keith Maslin, Stuart Hacker e Thomas Nagel. Sobre isso, especificamente, Miguel Nicolelis e o matemático francês Ronald Cicurel escreveram um livro inteiro.

A obra está disponível em inglês "The Relativistic Brain: How it works and why it cannot be simulated by a Turing machine" e traduzida também em português "O Cerebro Relativístico: Como ele funciona e não pode ser simulado por uma máquina de Turing" ----- aqui, você tem um dos cientistas mais acreditados no planeta atualmente a respeito de interfaces cérebro-máquina descreditando a possibilidade real do Jogo da Imitação proposto por Turing, de um ponto de vista puramente materialista, como o do próprio Alan ---- porém, não determinista e nem reducionista. É importante notar isso... também há uma determinada concepção de "ciência e tecnologia" implícita no pensamento de Turing a qual, me parece, é passível de crítica.

(Ainda que Nicolelis subscreva ao naturalismo biológico, seguindo John Searle, à época de Turing, me parece que ainda não seria empiricamente possível fazer considerações sobre propriedades emergentes em relação ao problema mente/corpo. E para aprofundar outras facetas dessa discussão, claro, temos Alvin Plantinga).

Mas, para resumir a referência ao Nicolelis em uma frase, eu diria o seguinte: A singularidade não está mais próxima, e permanece altamente improvável.

Outra coisa que achei muito fraca no texto é o que o próprio Turing chamou de "objeção teológica", e que ele mesmo descreveu em nota de rodapé como "possibly heretical", que mal gera discussão, dada sua rasura. Vou me restringir, nessa resenha, ao seguinte: ele o constrói a partir de uma premissa inválida para a discussão teológica, logo... o posto de "pior argumento" vai para esse, onde o autor sequer dispunha da pertinência mínima pertinência necessária ao (ar)razoável. Qual seja, e aqui, cito diretamente de novo:

but should we not believe that He has freedom to confer a soul on an elephant if He sees fit? We might expect that He would only exercise this power in conjunction with a mutation which provided the elephant with an appropriately improved brain to minister to the needs of this soul. An argument of exactly similar form may be made for the case of machines.

Novamente, não é uma questão, para Deus, de "posso fazer", como é para Turing, assumindo que Deus é amor, e Turing acha que com espaço de armazenamento suficiente, emoções, ou a imitação delas --- para ele, não haveria diferença real, pois se trata apenas de output--- também serão programáveis. Não, a questão aqui é que seres humanos têm alma porque é assim que são seres feitos à imagem e semelhança de Deus, e não porque almas poderiam ser concedidas aleatoriamente a algo que não o único ser feito à imagem e semelhança de Deus.

E a propósito da citação de Ada Lovelace: Com certeza, o trabalho dele seria impossível sem o dela, como "programadora" da máquina de Charles Babbage, mas Ada se restringiu ao empiricamente concebível. Ele a cita com óbvios respeito e admiração, porém, sua linha de argumentação é simplesmente buscar formas de extrapolar o que ela disse. Segue:

“The Analytical Engine has no pretensions to originate anything. It can do whatever we know how to order it to perform”. Essas são as palavras da Ada.. Alan achou pouco.

Este ponto é de particular interessante. O "gênio" de Turing nesse texto esteve justamente no fato de ele insistir na possibilidade das máquinas aprenderem a aprender. Se por um lado a frase de Lady Ada, em termos estritamente lógicos, abarca essa possibilidade, por outro, ela não autoriza que se perca de vista o humano e sua intencionalidade. Mas a chatice está justamente aí: a argumentação de Turing é toda voltada a desmantelar isso. Ele acha desejável máquinas competirem e "enganarem" seres humanos, no sentido da indistinção. Para ele, não importa o que, na verdade, seja uma coisa e seja a outra. Para ele, só importa função.

A mim, fica a impressão de "Quem não sabe, não precisa escolher", ou seja, um jeito fácil de escapar dessas problemas existenciais e éticos imbricados na questão. Jeito fácil também de restringir o tipo de liberdades que, muito especificamente, associamos a seres humanos.

Que isso é ótimo para ficção científica e péssimo para relações trabalhistas baseadas em exploração, já ficou bem claro em nosso tempo presente. Porém o ponto é que em ambos os casos, os imaginados e os que se encontram em franca concreção atualmente, são denotados aspectos disfuncionais e até doentios de nossas sociedades. Isso está adiante e além das considerações de Turing ou, antes, as considerações dele é que ficaram aquém dessas dimensões que são tão importantes, por serem as que que de fato têm a ver com a vida das pessoas depois que grandes escolhas societárias e culturais são postas em marcha.

E não, ele não estava apenas falando de informática. Ele estava propondo cosmovisão, ideologia e formas de sociabilidade... como se não estivesse, como se fosse só uma questão "neutra, objetiva, científica" de capacidade de armazenamento de dados. E é aqui que reside o problema, e é por isso que pessoas de humanas e suas problematizações são necessárias ---- *fica essa dica para a galera pragmática das tecnologias*.

**Já esta dica que se segue, é só para a galera mais inteligente das tecnologias, que sabem que lidam com meios de exercício de poder e, por isso, não fogem das difíceis reflexões éticas e se educam a respeito de outras coisas também**: Até onde minha ignorância permite ficar sabendo, sei que a Computação, hoje em dia, ao tentar se humanizar, parte de outras dimensões filosóficas que não o naturalismo mecanicista de Turing, a saber: 4E cognitition e Aesthetic Computing. Todas essas discussões levam em conta questões fato e valor metaignoradas por ele (se de maneira proposital ou apenas "espontânea" não caberia julgar).

Deixo as referências, para ninguém ficar parado no pós-guerra: Shaun Gallaguer, (https://www.youtube.com/watch?v=M7ghX... Chalmer, Clarck e Tallis (https://www.youtube.com/watch?v=rSngH...) e este paper https://www.interaction-design.org/li...

[Gavin · Rating 5 out of 5]

More scientific than it's given credit for: the claim about the Turing test was

Consider first the more accurate form of the question. I believe that in about fifty years' time it will be possible, to programme computers, with a storage capacity of about 10^9 [bits], to make them play the imitation game so well that an average interrogator will not have more than 70 per cent chance of making the right identification after five minutes of questioning. The original question, "Can machines think?" I believe to be too meaningless to deserve discussion.

Nevertheless I believe that at the end of the century the use of words, and general educated opinion, will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted. I believe further that no useful purpose is served by concealing these beliefs. The popular view that scientists proceed inexorably from well-established fact to well-established fact, never being influenced by any improved conjecture, is quite mistaken. Provided it is made clear
which are proved facts and which are conjectures, no harm can result.

We failed him on this specific timeframe, but it won't be too long (2030?).

[Shwetha · Rating 5 out of 5]

This original paper finally connects many disparate phrases and ideas I’ve been exposed to through sci fi stories, machine learning courses and pop science in general from the man who originated them. The father of modern computer science is a clear writer and uses simple language to describe his thoughts and arguments, and there were many arguments against it’s possibility, on the future of machine learning: implementation, use cases and philosophy; during the early years of digital computers.

He was able to predict the principle of evolutionary algorithms used today as well as structure a clear test to determine a thinking machine called “The Turing Test” popularly but he preferred to call the “The Imitation Game”. This article is easy to follow for beginners who’ve had at least a high school education.

Other works to read from the Bibliography:
Notes on the analytical machine by Charles Babbage, Countess of Lovelace’s Translator's notes to an article on Babbage's Analytical Engiro,
Godel’s incompleteness theorem,
Hartree’s Calculating instruments,
S.C Kleene’s General Recursive Functions of Natural Numbers,
Turing’s On Computable Numbers, with an Application to the Entscheidungsproblem,
G. Jefferson’s The Mind of Mechanical Man.

[Nicolas Name · Rating 5 out of 5]

Famous “Turing Test” is presented in this paper. Lots of predictions made by Alan Turing that turned out to be true (development of machine learning and AI around year 2,000, chess playing capability of machines surpassing humans, etc.).

His argument in favor of the Turing test stemmed from his approach to the question “can machines think?” or are machines conscious? His argument is that these words are not operation or something we can define, so he then went on to develop the Turing test, that it a machine can be programmed to behave in such a way as to be indistinguishable from humans that a human can’t tell apart the machine from the human, then we ought to view them as being conscious (since we apply a similar standard to other humans and whether or not other humans are conscious). He argues that whether machines are conscious, that the current standard being applied is that unless one physically experiences being a machine and conscious at the same time and has that experiential experience, then we can’t be sure, and that’s not really possible (outside of some science fiction story).

It’s a pretty fascinating discussion in the paper, that goes beyond programming and goes into the nature of consciousness itself.

[Leonardo]

Can a computer understand the meaning of a sentence? And how could we tell if it did? This is similar to asking “Can a computer think?” Alan Turing famously proposed to answer this by examining the ability of a computer to hold sensible conversations with a human (Turing, 1950). Suppose you are having a chat session with a person and a computer, but you are not told at the outset which is which. If you cannot identify which of your partners is the computer after chatting with each of them, then the computer has successfully imitated a human. If a computer succeeds in passing itself off as human in this “imitation game” (or “Turing Test” as it is popularly known), then according to Turing, we should be prepared to say that the computer can think and can be said to be intelligent. So Turing side-stepped the question of somehow examining the internal states of a computer by instead using its behavior as evidence of intelligence.

Natural Language Processing with Python Pág.367

[Bicho · Rating 3 out of 5]

Nunca me cayó bien Turing. Lo admito.
En fin, la cosa es que este ensayo de Alan M. Turing fue publicado originalmente en 1947 y ha pasado a la historia de la ciencia ya que se considera el antecedente de la IA. Es importante destacar el año, para ponerlo en su adecuado contexto.
La pregunta que da nombre al título era muy candente a mediados del siglo XX, cuando se comenzaban a vislumbrar las primeras computadoras digitales y atisbar de lo que serían capaces en un futuro no tan lejano.
Turing, en este artículo, no sólo explica su respuesta a la pregunta del título, sino que también intenta responder a todos los que, de alguna manera u otra, se presentaron como detractores, desde muy diversos ámbitos.

[Mira · Rating 5 out of 5]

I never imagined I'd appreciate and, more importantly, enjoy a work written by a mathematician, but I've been proved wrong! I also never imagined I'll encounter wit, sarcasm and conversational style in such a text, but here we are!
It's a wonderful paper, surprisingly relevant and accurate (as far as my literary-schooled mind can tell) despite it being written almost 70 years ago.
I wish I could tackle other writings by Turing because I do have a soft spot and tremendous respect for him but I feel that I won't get much of what he writes but who knows, I might do so eventually!

[Maxim Kavin · Rating 4 out of 5]

Эта работа Тьюринга — не книга, а статья в обложке. Конечно, Алан это заслужил своей деятельностью. В этой работе он приводит свои доводы о том, почему машины могут мыслить. Но он предполагал, что первая машина, которая пройдёт Тест Тьюринга, сможет это сделать в 2000-м году с объёмом памяти в ~125 МБ. На дворе 2021 год, и ни одна машина так и не прошла этот тест. Поэтому, к сожалению, мы пока не знаем — верна ли мысль Тьюринга, могут ли машины мыслить, а не имитировать поведение?

[heidi · Rating 5 out of 5]

Brilliant. I love that Turing spent a portion of his article meditating on whether or not a machine could develop its own free will, consciousness, soul, etc. Basically, exploring what it means to be human.

I'll be mulling over the points within this book for a long, long time. Such ramifications. Wow.

[Thomas · Rating 4 out of 5]

A super cool paper by Turing, about the idea of learning machines. It's hard to imagine that this was written 70 years ago, because much of what he writes describes the direction that has been followed by computer science since then. This really conveys the feel of a visionary man.

It's broadly accessible, I'd recommend anyone curious about machine learning to give a read to this seminal paper.

[LeelooVonCaster · Rating 3 out of 5]

El libro es en general bastante bueno, pero el capítulo de opiniones contrapuestas (hasta la objeción de Lovelace) me ha parecido tan aburrido que he dejado el libro parado casi un año.

Las teorías de Turing me han parecido muy acertadas y muy bien pensadas para la época, en concreto me ha gustado mucho su último capítulo.

[Hypia Sanches · Rating 5 out of 5]

Gente, que artigo incrível!!!! Alan conseguiu falar de uma infinidade de temas relacionados com inteligência artificial e aprendizado de máquinas num momento em que esses conceitos talvez nem tivessem recebido esse nome. Considerei uma leitura importantíssima pra qualquer pessoa da área de tecnologia, ou mesmo qualquer pessoa curiosa.

[Erik Perillo · Rating 5 out of 5]

I was surprised in a good way reading this. The ideas explored in the text are even deeper than I thought. Besides being full of humor and wit, the paper also gives a good idea about how brilliant and scientifically-minded Turing was.

[Mithren · Rating 4 out of 5]

Fascinating and easy to read. Could be used to introduce people on the subject as it is still relevant.

OBS: LOL at the bit about telekinesis and telepathy and how it is a solid argument against the efficiency and reliability of the imitation game.

[Benjamin Jordan · Rating 5 out of 5]

I’m not qualified to give this less than five stars. I do think that most attempts to make Turing’s imitation game more digestible are far more difficult to process than Turing’s own writing. This should be required reading for anyone in technology.

[Partheeshwar · Rating 5 out of 5]

I'm sure this book is linked with philosophy of mind. The first thing we should ask ourself is "what is thinking and what is mind"?. The book also deals with Mind-body theory and functionalism. Finally, the possibility of making an artificial intelligence to play an imitation game.

[Artem Lebedev · Rating 4 out of 5]

Super interesting how much Turing predicted. He disarms many arguments against machine superiority and makes it even more scary for me. His predictions hold true and I'm now convinced that our value is just s delusion - very soon machines will be more valuble for us than we ourselves.