MBZUAI president lays out bold vision to put UAE 'on map of AI superpowers'

The man leading the world’s first dedicated artificial intelligence research university has set himself a challenging target: to put the UAE “on the map of AI superpowers”.

Speaking to The National, Dr Eric Xing, a world-renowned computer science professor, who was appointed president of Mohamed bin Zayed University of Artificial Intelligence last year, outlined his bold vision for the pioneering institution.

“It’s very simple to say but very hard to execute,” he said.

“My vision is to really use this opportunity to turn MBZUAI into one of the major players, and use this as a platform to train talent and the workforce for the local economy and for the community.”

Prior to joining MBZUAI in November, Dr Xing was associate head of research at the machine learning department at Carnegie Mellon University in the US, one of the leading AI research institutes in the world.

He said his first priority for MBZUAI was to develop a “critical mass” of research output and impact in the shortest amount of time possible.

“I don’t want to be the one who boiled the ocean by doing everything, and end up mediocre in all dimensions,” he said.

Instead, the university’s curriculum is focused on two areas: machine learning, which provides the mathematical foundation of AI, and computer vision, which takes machine learning a step further to identify and analyse images and videos.

Admissions are also now open for its third course, Natural Language Processing, which allows computers to communicate with people using everyday language, beginning in Spring 2022.

Over time, the plan is to add departments such as healthcare, energy and industry, Dr Xing said. “Those areas where there is a good synergy and a strategic priority in the nation for us, the UAE, then we want to tap into our ecosystem and to help our community to benefit.”

He has done this sort of specialisation development before, as a founding director of the Centre for Machine Learning and Health at Carnegie Mellon and the University of Pittsburgh.

He also said he hopes to bring AI “out of the ivory tower” and put it into practice in the UAE.

“Usually with research universities, the only goal is to produce more researchers who will be entering other universities as professors,” Dr Xing said. While that remains important for MBZUAI, he said he would like graduates to go on “to be engineers, executives and government officials in our community”.

“We’re in a stage of AI development where there is still a lot of fear,” he said. “It’s like when the automobile was invented. People were scared and they wanted to keep using a horse and buggy.”

But building a pipeline of graduates with expertise in the field may help change perceptions about cost and risk and increase the influence of AI in the private sector and government, particularly locally, he said.

Despite the US and China leading the world in terms of level of investment and retaining talent, Dr Xing pushed back on the notion that these two nations are dominant.

“There are many many players making contributions,” he said.

“Google of course is a great practitioner. To use their big infrastructure and resources to train very impressive, big models and using big data in an impressive way.

“But on the other hand, if you look at the people who work in Google to build those models, they actually trained from [Carnegie Mellon], MIT and other top schools. So there is a symbiosis.

“It’s not like the solo aria in opera, it’s literally the chorus that is needed.”

In the battle against Covid-19, he pointed to efforts in the UK to work with Oxford, Cambridge and Imperial College London to develop AI models to help address challenges.

Yet AI fell short in its pandemic response, in terms of public information.

Dr Xing, who served as a visiting research professor at Facebook, said he sees enormous potential for AI to help social media companies more quickly identify the spread of false information and accelerate messaging that can improve public health, such as the adoption of mask wearing.

“AI will be a great technical resource addressing these problems,” he said.

“AI could help because AI is very good at collecting and analysing big data. You actually quantitatively understand what it takes for the public to internalise a policy, or to receive a particular piece of communication, and how soon they will be psychologically more comfortable.”

This month, MBZUAI admitted its first class of students, 13 pursuing doctorates and 65 earning their master’s degrees in machine learning and computer vision.

The deadline for applications for the autumn 2021 semester is April 15.