Prof. Ajay Agrawal: 3 steps for fostering AI talent
Professor Ajay Agrawal is from the University of Toronto’s Rotman School of Management, where the future that AI is forging is being studied from an economics perspective. He has 3 strategic steps that companies should be following at the moment in order to foster AI talent.
Professor Agrawal is the leading expert on AI in economics, and his book “Prediction Machines — The Simple Economics of Artificial Intelligence”, co-authored with two others, carefully explains the importance of AI in business and the mechanisms of its impact. The book doesn’t just target the executive and management levels but everyone in a business, from new employees to middle management. It also answers many of the questions we all have concerning AI.
When Professor Agrawal visited Japan in March this year, BeautyTech.jp was there to ask him on the latest developments since the launch of his book in 2018.
His extensive knowledge of both business and AI comes from being the founder of the startup-supporting AI business creation program Creative Destruction Lab (CDL). It was founded in 2012, and by 2017 was, for the third consecutive year, the program with the largest participation of AI-related startups in the world, producing stock value worth a total sum of US$2.07 billion.
3 steps that sum up how to foster AI talent
On releasing his book “Prediction Machines”, Professor Agrawal reveals that he was called on by many corporations to discuss their business strategies. He found that even though they had come to understand the importance of AI, the corporations didn’t know how to go about making changes to their company. This is because implementing and utilizing AI within a business requires thinking in terms of the organization as a whole, including the employees.
When writing his book in 2017, Agrawal saw the effects brought about by prediction machine AI as being represented in five stages, in a layered pyramid structure. At the base is “Prediction”, followed upwards by “Decision Making”, “Tools”, “Strategy”, and with “Society” at the top. “I thought companies would be able to come up with the answers themselves by thoroughly contemplating these 5 elements, however, what we also needed was a method for changing companies.” Change can be difficult — the rejection of new things or regretting not being able to change despite wanting to are issues universal to many cultures around the world. “Nevertheless,” Agrawal maintains, “if you don’t change you’ll lose to the competition.” Thus the professor advocates the following essential 3 steps for incorporating AI talent into companies.
1. Teach AI to current employees
The past several years have seen the appearance of many new AI courses that are available for free and are of high quality. This step begins by making use of these courses to teach current staff about AI and increase their understanding. Although obtaining top AI talent may not be easy, “the most important thing is to think through what the effects of AI are for your company and take action”, says Agrawal. To do that, you need employees who can make optimal decisions in consideration of the organization’s mission and purpose. So as a first step it’s important to increase the number of people in-house who have knowledge of AI.
2. Be more aware of taking on the global market and actively invest in education
Professor Agrawal points out that the procurement of global talent will likely become more feasible as “before you know it the day will come when AI will make the local language barrier disappear.” This is another reason, he says, that “if you think AI personnel are essential you ought to be actively investing in education.”
One case study is Toronto, Canada, a city which has yielded a great number of top global-class AI researchers thanks to the Canadian Government earnestly investing in AI education. The enterprise has had its ups and downs, but with the initiative of the Canadian Institute for Advanced Research (CIFAR), the below four goals were set as an embodiment of the AI strategy, and in 2017 a budget of CAD$125 million was attached.
● Increase the number of outstanding AI researchers and post-graduate students in Canada;
● Strengthen cooperation between the AI research centers in Canada’s three major cities of Toronto, Montreal and Edmonton;
● Become a global pioneer for the economic, ethical, policy and regulation aspects of AI; and,
● Support the AI research community.
(Source: CIFAR Pan-Canadian Artificial Intelligence Strategy)
Although the bureaucracy and slow-moving tendencies of governmental and academic institutions have been obstacles, the search for new ways to freely conduct research and development unrelated to industry, government or academia and cultivate outstanding AI talent has resulted in the launch of a new independent institution. Support came in the form of CAD$40 million from the Canadian Government, CAD$50 million from the Ontario provincial government, and a total of CAD$80 million from the private sector, enabling the establishment of the Vector Institute in March 2017, a non-profit organization specializing in deep learning and machine learning.
3. Be committed to tackling diversity
Professor Agrawal’s theory maintains that “if companies are seriously thinking about obtaining AI talent they ought to place more importance on diversity. A major company can easily partner with a university and train up to ten thousand students into AI-capable employees. However, it is by actively hiring more women, and also foreigners, that the competitiveness of a company can be fostered.”
In the US only around 26% of IT engineers are women, however, in Japan, it’s around 19% (data from p. 28 of the 2018 Japan Information Technology Services Industry Association Basic Survey). Around the world, there are companies that are realizing greater capabilities by actively hiring female engineers. One such company is Sephora, a multinational chain of cosmetic stores. Their proactive hiring of female engineers has led to their technology staff becoming over 60% female, and this has brought about certain merits. Because the female engineers are closer in age to Sephora’s main consumer demographic and also have a high interest in beauty and makeup, staff overall has a stronger alignment with customers for when thinking about how to utilize technology.
All companies are facing the issues of how to attract talent amidst a tight pool and how to train existing staff to be more competitive. However, it’s not difficult for companies to get serious about investing and cooperating with AI-teaching universities to train existing employees and students into AI-capable personnel. What’s most important is for companies to get serious about AI and invest in it.
Text: Ching Li Tor
Original text (Japanese): Yukari Akiyama
