A constant focal point of contemporary news and media, artificial intelligence is playing an increasingly significant role in various sectors of life; from healthcare and transport to stock trading and education. According to Merriam-Webster, artificially intelligent machines are defined by their capacity for “simulation of intelligent behaviours” normally associated with humans, while the English Oxford Living Dictionary notes that artificially intelligent machines can “perform tasks… such as visual perception, speech recognition, decision-making and translation.” They carry out tasks to facilitate our ease of life and reduce human labour; skills we will explore in relation to academic research in universities.
Undoubtedly, artificial intelligence is revolutionising university research and helping students cope with the challenges and hurdles of higher education. For example, Iris.ai, the world’s first artificial intelligence science assistant, supports academic research by sifting through the wealth of journal articles available and finding relevant articles for research questions.
Like other forms of artificial intelligence, Iris.ai reduces human labour and conserves time, with the company claiming that it can solve your research problems 78% faster (without compromising quality) than if you were carrying out the tasks manually. Drawing on the publication rate of academic articles highlights the usefulness of this tool; 2.5 million articles were published in English journals in 2014 alone.
With the sheer amount of information available, we simply do not have enough time to navigate them all and manually find the most relevant articles, which is where Iris.ai comes in. Iris.ai’s ‘mirroring of human intelligence’ is evident in the way it carries out its research; it has contextual understanding of the question which allows it to build the literature review. Unlike typical search engines like Google Scholar, the program looks at the themes within an article (not just the title and keywords), highlighting connections and trends across papers and even disciplines. This highly advanced function produces more specific and relevant articles, facilitating research.
Semantic Scholar is very similar in this regard, as it advertises itself as being able to “cut through the clutter”, and provides a ’smart’ search engine for journal articles. It aims to help us ‘keep up’ with the things we are discovering daily and the various detailed academic research being produced continuously. There is clearly a recurring theme of artificial intelligence being programmed to make mundane research tasks easier and save time. Semantic Scholar’s potential is limitless. Oren Etzioni, the head of the Allen Institute for Artificial Intelligence which is responsible for the product, plans to expand the program into a “scientific assistant which can suggest new hypotheses or new directions for research,” opening up new possibilities for scientific research.
Artificial intelligence can also be used to uphold standards in academic writing and research. It is used by publishers and universities to detect plagiarism and misuse of statistics, ensuring that new articles that are being published are of a high quality. This also ensures that when students are researching, they are presented with articles which are checked thoroughly and the information is reliable and accurate.
Bolton College has demonstrated pioneering advancement in relation to artificial intelligence. Just earlier this month The Bolton News credited the College as being the “only further education establishment nationally to actively use AI to support students’ learning.”
It has achieved this through a software known as ‘Ada’ (also referred to as ‘Ask Ada’) which is their institution’s equivalent of Siri or Alexa. Ada is a “cognitive assistant” that allows students and staff to gain information in a more accessible manner. It has garnered widespread international interest and was viewed by visitors from Canada and America representing the IBM Watson Conversation.
The concept of artificial intelligence has also caused fear, panic and hysteria as scientists warn of the potential replacement of humans by machines as their intelligence supersedes our own. This has also been reflected in popular media and films such as The Matrix trilogy, Terminator franchise, and Avengers: Age of Ultron.
However, it is important to note that these machines imitate and mirror human behaviour. Forbes notes that they are “human-like rather than becoming human,” and the Encylopaedia Britannica notes that “as yet no programs that can match human flexibility over wider domains” exist.
As long as we keep programming artificial intelligence machines, we dictate what they are capable of carrying out and we remain in control. These machines are not able to act independently without instructions from humans.
For example, in the case of Iris.ai, a team of 10,000 people is responsible for directly training the software. People read the abstracts of research articles and manually tell the program what the core concepts are.
Therefore, the ‘contextualised understanding’ attributed to the program is not its own inherent knowledge but one that is reliant on a human’s interpretation and conceptualisation of the research.
Furthermore, it is inevitable that many roles and tasks will be carried out by machines instead of humans.
This said, it is worth remembering that artificial intelligence creates several new job vacancies that did not exist before.
There are a great many roles associated with the field and huge prospects for careers in computer science.