Machine learning is fast advancing to be able to teach humans, making human command in the decision-making end role more important, according to Professor Eric Xing, President of Mohamed bin Zayed University of Artificial Intelligence (MBZUAI).

AI technology has rapidly progressed over the last decade from assistance and customization based on various algorithms to management of vast data, Xing said as he spoke on the evolvement and direction of AI Thursday on the last day of the 23rd World Knowledge Forum in Seoul.

AI machines currently are capable of simple tasks, like answering to questions with already obtained data, but they will be able to make predictions and decisions by making inference from unseen data, Xing projected. Ultimately, they will be capable of teaching humans, he added. AI is now more sophisticated with more advanced language model and customized recommendations backed by various algorithms.

AI has allowed use of more advanced mathematics in digitalization, which in result has led to the creation of various new technologies that would be very useful in the next 10 years, said Eric Daimler, Co-founder & CEO of Conexus AI, who also attended the session and discussed the progress of mathematics with the emergence of AI.

AI should be more upgraded so that it can learn by itself and make complex predictions from both trained and unseen data, the experts stressed.

With AI’s more diverse and massive data, people will be able to make a better decision in various social issues such as climate change and supply chain bottlenecks, Xing said, adding humans must have the key in making a right and safe decision.

By Kim Geum-yi and Jenny Lee

[ⓒ Pulse by Maeil Business Newspaper & mk.co.kr, All rights reserved]