The Yesterday

The father of theoretical computer science, Alan Turing, published his groundbreaking research paper, ‘Computing Machinery and Intelligence’ nearly 75 years ago. In this paper, he made a now renowned prediction, that computers with 100 megabytes of memory would be successful in convincing a tester that they were human themselves through the quality of the conversation they would be capable of maintaining.

The goal of this paper was to sidestep the deeply existential questions that often arise when pondering the fundamental mechanics of consciousness, sentience, sapience and intelligence. It does this by simplifying them, attempting to tackle the problem as a function of the extent to which a machine is capable of imitating a human being. While ‘The Imitation Game’ or Turing Test disguises itself as research interested in offering a test to determine whether a machine is truly intelligent or not, in reality, it is a thought experiment that makes the argument that consciousness isn’t as unsolvable as most believe and that looking, walking and quacking like a duck is indistinguishable from being the duck.

Further discourse has yielded the most prominent counter-argument, ‘Chinese Room’. This thought experiment puts a librarian who does not speak Chinese in his library with an infinite array of books that each alphabetically contain a generic answer to every possible Chinese question. A Chinese speaker then passes a question chit from outside the library which the librarian promptly answers via chit utilizing his infinite book supply. The Chinese man outside the library may now be convinced that the librarian speaks fluent Chinese despite the far bleaker reality.

Criticisms of the Chinese Room argument posit that while the librarian himself may not know the language in an isolated environment, if the library and librarian are perceived as a unified system, they do/it does speak Chinese (more precisely, it is capable of answering any conceivable Chinese question asked via chit, with the library providing the actual answer and the librarian doing the work to read the question chit, find the question within the library and write the given answer on the chit and return it to the Chinese man outside.)

If this flow of logic irks you, dear reader, ask yourself this; What component of you really “knows” English? Certainly, if your arm was chopped off, you’d still speak English. What about your heart? Well, you wouldn’t really be doing much of anything without a heart so it’s a leap to claim that your knowledge of English is stored in the heart. Let’s head northward and interrogate the brain, but where in the brain? With some knowledge of neuroscience, we could narrow it down to the left hemisphere but there’s a limit to how physically specific we could get.

Language is a relatively delocalized phenomena, it originates from outside you and exists separate from you so to delve into the heart of the hard problem of consciousness, we must discuss what Turing opted to specifically avoid. The most likely reason he picked conversational language as the perfect dependent variable is that more localized experiences have very little scope for physical manifestable observations. To understand this idea, consider the Chinese Room but instead of asking a question in an unknown language, the external party must test the librarian’s EXPERIENCE of a colour, a taste or a smell. You might be able to outline the precise wavelength of electromagnetic radiation that lilac reflects, the precise gustatory receptors that the chilli chutney accompanying a crispy samosa activates or the precise chemical reaction taking place within your olfactory epithelium as your mom swishes the Agarbatti to fill your room with it but you will never be able to explain how either of those things FEEL to a man disaffected by colorblindness, ageusia or anosmia.

This deeper feeling or experience of a physical thing, called Qualia, is entirely separate from the describable physical properties and objective nature of the thing.

The Today

Recent developments in large language models bring us ever closer to that 100% Turing Test success rate and beyond. In an era of development as rapid as it is, with OpenAI introducing Sora, a gigantic leap in AI generative video, one that is evidently threatening a whole new assortment of jobs; filmmakers, animators, editors and many more professions in visual media are now joining the list of occupational fields likely to be integrated with Machine Learning models, ushering generational automation transformations.

With all this uncertainty, it might have helped you, as it has helped me, to peer back in time and understand the history of the theoretical discourse surrounding the development of an Artificial General Intelligence, through the lens of understanding human intelligence. What separates the statue of a human, monkey or worm made from every ingredient that it is composed of down to the atom, from the human, monkey or worm itself? Is finding out important? Why should we care?

The Tomorrow

Leading experts in data science, philosophy, psychology and neuroscience now believe that unless all the silicon on planet Earth fizzles away before the Bezoses and Musks of the world are able to mine a stolen asteroid to yield a fresh trillion tons of whatever magic goes into the GPUs that large models run on, the future will get consistently more artificially intelligent than the past. That means that there will certainly be a day when a large machine-learning model can engage in a conversation with a human with errorless precision, but that doesn’t seem to shock us as much as it might have a layman reading Turing’s paper in 1950. What is genuinely shocking, is that we landed on the moon before we invented hair gel. Humans, when there are enough of us dedicated to the right causes and the welfare of the species, can pull off unbelievable feats. Will a day come when a man who bleeds motor oil, scans rather than sees and records rather than hears will be able to truly feel the pride that we do upon hearing the words in that croaky, distorted, unclear yet hope-filled voice, “That’s one small step for man... one giant leap for mankind”?

Well, it hasn’t yet, and if it does, until it does, I will find comfort in the knowledge that the genius who won Britain the most important war in human history grossly underestimated the complexity and beauty that lies buried within the depths of the most compelling mystery of the 21st century; What makes a monkey?

.    .    .