There is no doubt that we live in the time of smart devices. From the simplest everyday items, like a coffee maker which can be programmed to have your coffee ready the moment you get out of bed, to the smartphone in your hand. When people talk about artificial intelligence (AI) though, their minds go a lot further than that. They seem to think of evil robots with the desire to take over the world. But is that even possible? And what exactly can be done with AI?
AI is the term computer scientist John McCarthy used to describe “the science and engineering of making intelligent machines”. There is no denial that computers nowadays can perform a large number of tasks better and faster than any human could and in fact they are improving at rates that humans simply can’t match. The immense power of computers to analyze large amounts of data, identifying patterns and finding connections, has an obvious application in fields such as the stock market and economics, but it has also been invaluable in biomedical sciences. Ever since whole genome sequencing jumped from being an expensive 10-year long experiment to a relatively cheap everyday technique in the lab, scientists found themselves with a plethora of data in need of analysis. As a result, high-power computers and machine learning (ML) methods have found new applications in biomedicine as a way to analyze complex datasets to predict the outcome of deadly diseases.
This can be an invaluable diagnostic tool and the number of studies using ML techniques has increased over the last decade, however there is still a major caveat. Trying to identify the patterns and key factors involved in a disease, such as cancer, is like trying to reverse engineer a highly complex system whose detailed functions are not yet fully understood and are certainly not always based on Boolean logic, but rather manifest in shades of grey. Additionally, the applicability and precision of ML techniques in interrogating diseases depends a lot on the quality of the collected data, which in turn means that if biomedical scientists plan to include such methods in their experiments there will need to be guidelines and quality criteria if they are ever going to be useful as a diagnostic/therapeutic tool.
In the meantime, the current definition of AI lacks one very important aspect of human intelligence, namely the cognitive skills. Naturally the notion of transferring consciousness to a machine has gathered the interest of different scientific fields and huge strides of progress have been made. From Google’s DeepMind self-learning algorithms to the development of an artificial DNA-computer which can give the correct answers to a series of questions an interesting question arises: can we grow AI? Spoiler-alert: we probably can. Recently a passionate computer scientist (Toni Westbrook) did just that. Synthnet is a virtual brain, a virtual neuronal network formed by virtual DNA, which can learn through trial and error process, essentially recreating intelligence. Also very recently a non-invasive brain-to-brain interface (BBI) was published whereby a functional link between the brains of a human and a rat was established through a computer, successfully transferring thoughts into motions from one brain to the other.
It is clear that AI is the new kid on the block, a kid that is growing at a really fast pace and it is also safe to say that its place in the future is guaranteed. However, when facing a better, faster and smarter computer it is not good enough only to ask: what can we do that a computer can’t? We need to ask what activities will humans always insist be performed by other humans. The answer to this question is rather simple: relational tasks, tasks that include empathy, function in a group and social awareness. Until we are able to solve the ethical dilemma of algorithmic morality or to put it simply: ‘what should a self-driving car be programmed to do in the event of an unavoidable accident’, or until we reach a point where we can fully transfer consciousness to a machine, we humans are going to stick around for a while.