Nowadays, we hear more and more about artificial intelligence and its new applications. Intelligent machines and programs recognize human speech, translate texts into other languages, drive cars, play chess with humans and even create art. Applications of AI (artificial intelligence) of varying degrees of complexity can be found today in the most unexpected areas of life.
One of the newest projects using artificial intelligence (in this case neural networks) generated short stories imitating the literary style of Polish writer Olga Tokarczuk. Artificial intelligence was “fed” with the content of 9 novels of the Nobel Prize winner and it created its own work inspired by them. The story was published with the consent of the writer and read by human actor Janusz Chabior. We will probably have to wait a little longer for the day when artificial intelligence creates its own unique style and deserves a literary Nobel Prize, but it must be admitted that even this short story is quite impressive and its AI creator probably exceeds the writing skills of many people.
Another project, created by the Canadian start-up BlueDot, warned its clients about the possibility of a COVID-19 outbreak near a Wuhan market nine days before official warning from the World Health Organization. Using machine learning and natural language processing, media reports and medical bulletins published in different languages were analyzed to identify the risks and alert the Canadian health authorities. AI has also identified the countries most at risk of spreading the virus based on an analysis of airline ticket data. Earlier, BlueDot had also predicted the arrival of the Zika virus in Florida in 2016 six months in advance, as well as the spread of the Ebola outbreak outside West Africa in 2014. Early detection of epidemic threats with tools such as BlueDot could prove crucial in suppressing them at an early stage and thus saving the lives of thousands of people.
We already know what artificial intelligence is capable of, but what is it actually? What distinguishes intelligent solutions from ordinary algorithms? The very term “artificial intelligence” was first used in 1956 by the American computer scientist John McCarthy. The role of artificial intelligence is to carry out functions of the human mind that elude the usual algorithmization, such as decision making in the absence of all necessary data, playing logical games or the analysis and synthesis of natural languages. AI involves use of techniques such as machine learning or neural networks, which are modeled on biological networks of neurons. However, the very concept of artificial intelligence remains vague and the delineation of its boundaries raises doubts. Artificial intelligence can be understood as a simulation of human intelligence and decision-making processes undertaken by humans to solve specific problems. However, there are cases where AI exceeds human mental abilities in certain areas and can be used to deal with previously unresolved problems. According to the degree of its advancement, we can distinguish weak artificial intelligence and strong artificial intelligence. Weak artificial intelligence is characterized by the fact that, thanks to the built-in self-learning algorithms, it works autonomously. This means that it is not subject to human control, or that the human supervision is limited to exercising the ex-post control. On the other hand, strong artificial intelligence should also manifest self-cognitive skills. Currently, however, it is only a theoretical concept and practical applications of artificial intelligence, with which we deal today, fall into the area of the so-called weak artificial intelligence.
Will artificial intelligence change the world we live in? We have no doubt that it will. The change is happening today, and the AI development is constantly intensifying and accelerating. Like any great change, it will also be a challenge for lawyers – both those who create legal regulations concerning new technologies and those who will have to deal with the legal aspects of artificial intelligence in practice. But this will be discussed soon in the next part.