Issue link: https://maltatoday.uberflip.com/i/1463329
OPINION 31.3.2022 Alexiei Dingli Prof Alexiei Dingli is a Professor of AI at the University of Malta and has been conducting research and working in the field of AI for more than two decades, assisting different companies to implement AI solutions. He forms part of the Malta.AI task-force, set up by the Maltese government, aimed at making Malta one of the top AI countries in the world O ver the past few years, we've seen a surge in the use of Arti- ficial Intelligence (AI). Many de- scribe this period as an AI summer with researchers actively churning innovative solutions and businesses benefitting from them. Below, we will look at the global state of AI to understand its effects in the coming years. Global investment is exploding and the amount of money pouring into AI is mind-boggling. Just think that this in- vestment mushroomed from 18 billion Euros five years ago to 160 billion Eu- ros today, a massive increase of almost 900%. e absolute majority (54%) of this funding comes from private invest- ment, and a big chunk of around 41% results from mergers and acquisitions. It is also interesting to note that there's a lot of consolidation between compa- nies, and the number of AI startups is decreasing. Countries are also actively working together on advancing AI research. e USA and China are the top two collaborating countries, with 300% more collaborations than others. Even though their international relationships sometimes seem strained, this did not discourage their research teams from working together. Apart from these two, we also find active collaborations between the United Kingdom, Germa- ny, Canada, Australia and France. One way to measure innovation is pat- ent filing. A patent is a right granted to an inventor by a government certifying that the idea is his for a limited time. e system seeks to encourage inventions that are unique and useful to society. China files around 87,000 AI patients per year, but only 1,400 get accepted. e US follows with 19,600 patents, out of which 9,450 get approved, and trail- ing, we find the EU with 4,800 patents filed with only 1,800 patents accepted. From this simple statistic, it is evident that the EU needs to do much more if the content wants to ascertain its dom- inance on AI research. Hence why, in her State of the Union for the current decade, the President of the European Commission, Ursula von der Leyen, de- clared this as the digital decade of Eu- rope. She mentioned three significant priorities: the data, emerging technolo- gies, particularly AI, and the digital in- frastructure of the continent. We are also experiencing a shift in technologies. In the past few years, we've seen the rise of Computer Vision. It is a subfield of AI that uses digital cameras to identify objects or perform facial recognition (to give a few exam- ples). However, this technology seems to have reached a plateau in recent months. e reason is that once Com- puter Vision completes its task, the AI cannot perform advanced reasoning us- ing that information. Let's look at an example. If a security camera identifies two people, it cannot infer additional information, such as whether they are taking a stroll together or if one of them is mugging the other. So the next step for AI is to move be- yond simple recognition and more to- ward the semantic understanding of real-life situations. Large companies such as Microsoft, OpenAI and others are already working on this challenge. ey created massive AIs capable of writing text at the same level as a human. When OpenAI creat- ed called GPT3, Elon Musk (who owns the company) refused to release it, say- ing that the model was too dangerous! Today, we have models which are many orders of magnitude more powerful. Notwithstanding this, these systems find it hard to understand and reason on text, especially when the complexity of the writing increases. Another area proliferating is AI ethics. Interest in the area grew by more than 400% in the past five years. Where- as conferences were traditionally the realm of academics, recent events have shifted towards private-sector involve- ment and governments. is interest is not surprising since AI systems are now entering mainstream use, and as a re- sult, ethical considerations are becom- ing more critical. To ensure that this momentum is maintained, it is essential to have a pipe- line of AI engineers. Many institutions have experienced a growing interest in AI-related courses in the past years. However, a look at the gender distri- bution quickly reveals that very few women opt for a career in AI. is situ- ation is a massive problem because the field needs a broader representation of professionals. Of course, this is not something new and keeps on popping up every year. Maybe it is high time that countries take action to resolve it once and for all. And they have to start long before people get into Universities. Finally, it seems that most govern- ments are finally waking up to the re- ality of AI. e number of bills passed through parliaments increased by 900% in the past five years, with Spain, the United Kingdom and the USA in the lead. Some allocated funds to bootstrap new AI initiatives, while others enact- ed regulations to mitigate risks. Gov- ernments are looking at AI as the tool which will transform their country in the coming decades. Of course, there's a lot of work left. But the future does look rosy, and all of us can expect significant changes in the coming years. The state of AI in 2022