What is AI as the core technology of DX? -Forbes JAPAN DX SUMMIT Report!

The online distribution event “Forbes JAPAN DX SUMMIT” was held on September 29th.

In the Closing Keynote Session that concludes the event, the main theme was “AI x Business ──What is AI as the core technology of DX?”

In this session, Mr. Yutaka Matsuo, Chairman of the Japan Deep Learning Association (hereinafter referred to as JDLA) and Mr. Andrew Ng, founder of DeepLearning.AI, took the stage, and Mr. Takashi Takekawa, CEO of zero to one and JDLA Human Resource Development Committee acted as moderator.

In 2012, when he was a professor at Stanford University, Mr. Ng released a computer science course online. For over 10 years, he has been a world leader in

In addition, Mr. Ng founded DeepLearning.AI in 2017 and launched an online course “AI for Everyone” for non-engineers in 2018.

JDLA will launch a new course “AI for Everyone” in May 2021 as a special edition that adds content for Japan produced by JDLA and lecturer Matsuo to the regular “AI for Everyone” delivered in English. I was. In June, one month after the launch, the total number of enrolled students exceeded 10,000, making it a popular course.

In this way, we will report on a part of this session, which was presented by Mr. Matsuo and Mr. Ng, who are spreading the importance of AI utilization and human resource development to the world from the forefront.

*In this article, some of the lecture contents have been omitted and expressions have been changed.

Session: What is AI as the core technology of DX?

Current status of DX | Improvements in AI technology play an important role in promoting DX

At the beginning of the session, Mr. Ng and Mr. Matsuo each gave their impressions on how DX has become popular as an important keyword in Japan in recent years.

Mr. Matsuo touched on the fact that the public and private sectors are actively working on DX in Japan .

Mr. Matsuo: The Japanese government launched the Digital Agency on September 1st of this year, and is trying to promote the digitization of administrative services and the efficiency of the core administrative system. In the past few years, many manufacturing companies, including Toyota and Panasonic, have taken notice of the possibilities of AI from an early stage and made changes, but recently, the movement of DX has accelerated in financial and insurance trading companies as well.

Mr. Ng explained that the use of data will become an issue going forward.

Mr. Ng: In the United States, many companies have talked about DX so far. I think many companies and industries have been working on this concept for years. From my point of view, the current situation of many companies is that the IT infrastructure is steadily being developed. But I don’t think there will ever be enough IT infrastructure.

The challenges faced by many companies in the United States, Japan, and around the world are how to collect data and how to use it, now that IT and data infrastructure have been established to some extent. I’ve heard some CEOs say, ‘I have two more years to build an IT infrastructure, and when it’s done, I’ll work on AI,’ which almost always leads to failure. Because the IT infrastructure will never be enough. It is desirable for most companies in the United States and Japan to first use the data they have now to produce results that create added value, even if they are small, and then continue to improve their IT infrastructure based on what they have learned.

The discussion then shifted to the role of AI in driving and accelerating DX within companies and organizations.

In response to Mr. Takegawa’s question whether he thinks AI is the core technology of DX, Mr. Ng replied, “It is the core technology.” He said that with the ability to process data, there are now opportunities in every large industry to create significant value using AI .

Mr. Matsuo also said that AI is a technology that greatly expands the possibilities of DX. Giving examples such as robot operation and natural language generation, he explained that ” improving the accuracy of AI technology plays an important role in promoting DX .”

Google was initially skeptical of deep learning

Next, Mr. Ng, who has helped launch and project various companies so far, was asked about an American company that has succeeded in DX, and he cited the global IT company Google.

Ng: It’s easy to forget, but even Google, which has great technology today, had to do an AI transformation back then. When I started the brain team at Google many years ago, everyone was skeptical about deep learning, including people inside Google.

I remember working on speech recognition in one of my first projects. I was working on the Google Speech team to improve the speech recognition system, but it wasn’t Google’s most important project because it wasn’t Internet search or advertising at the time. But with the quick win, more people within Google began to understand the power of deep learning.

The success of the first allowed me to plan a second important joint project. The purpose was to more accurately identify the location of private houses and buildings on Google Maps by reading house numbers from Google Street View images using exterior character recognition. When the second project turned out to be a success, serious discussions began (with Ng’s research team) with the online advertising team, which clearly had a direct impact on the bottom line.

Even Google, now one of the dominant and influential IT companies in the world, once did not emphasize AI as a core technology. While promoting projects within Google, the importance of AI came to be recognized, and now it is leading the world’s AI services with its advanced AI technology.

According to Mr. Ng, what can be seen from the Google case is that even today’s excellent AI companies have had a process in which their organizations learned and adopted technology . He said that even industries that were thought to be far from digitalization, such as corporations and media companies, are embracing AI.

Importance of AI human resource development｜The role of non-engineers is essential to demonstrate the potential of AI

In the second half of the session, we also talked about developing human resources to lead AI and DX projects.

Mr. Ng and Mr. Matsuo are both people who have put effort into training AI human resources. Mr. Ng also talked about the aims and use cases of the online course ” AI For Everyone ” for acquiring AI literacy, which was launched in May 2021 by JDLA in 2018 . .

When asked about his journey as an AI educator in the last 10 years, Mr. Ng said, “Hundreds of thousands of people around the world have devoted their time to learning about AI, building AI systems, or building AI systems. I am impressed that you are strategically acquiring knowledge about AI.” After evaluating the spread of AI knowledge, he answered as follows about his own efforts.

Mr. Ng: A very creative digital medium has put thousands of employees through online courses. The following year saw many innovations and applications for AI machine learning projects. There is also the case of the president of a country actually having everyone in his cabinet take “AI For Everyone”, and over the next few years the government began to set sensitive laws.

So whether it’s a company manager, executive team, CEO, country head, or a company’s entire workforce, creative ideas are born that will drive people to learn and apply new technologies to their business. increase.

On the other hand, in Japan, Mr. Matsuo is leading the education of AI human resources as a pioneer. Mr. Matsuo started giving lectures on deep learning at the University of Tokyo in 2015, and many of his graduates have launched successful AI startups.

Mr. Matsuo: I thought it was important to create some kind of qualification for the sound development of the industry. We established two qualifications, the G test and the E qualification , and also launched a large community called ” CDLE ” in which 40,000 successful applicants participate. This number is not many when considering the whole of Japan. So, in order to further expand the base, we have released the Japanese version of Andrew’s “AI For Everyone”, and it has attracted a lot of people so far.

Mr. Ng is the developer of “AI For Everyone,” and has a track record of about 700,000 people taking the course around the world. When asked about the intention of expanding the target of education not only to engineers but also to non-engineers, that is, to “everyone,” Mr. Ng replied,

Mr. Ng: In order for AI to maximize its potential, we need engineers, but we also need non-technical people to play a major role . With AI becoming so pervasive, almost everyone is affected by it. That is why it is important for business leaders and those in various functional roles to understand what technology can and cannot do.

I believe that business leaders who understand AI will be able to make better decisions, and given that AI exists in the process of DX in any industry, I believe that AI will help leaders and individuals with various roles. You are required to understand what is important . Having a broader knowledge base will help us make better decisions as a society.

Ms. Ng, “Realization of AI is the responsibility of each citizen” Ms. Matsuo, “A willingness to change”

In the final stage, Mr. Ng answered the secrets from successful cases around the world about what Japan needs to do to accelerate DX from now on.

Mr. Ng: The first is education. The best first step was to have an imaginative brainstream with “AI For Everyone”.

The second is community. You can’t learn anything by yourself. Therefore, it is important for organizations such as JDLA to gather people from all over Japan who are working on deep learning and to openly share ideas.

Third, I would like companies to see the present as an opportunity and work quickly. I have seen many companies aim big and fail. Remember that the purpose of the first project is to learn from you for the next bigger project. I think it can be a valuable experience, and sometimes bottom-up opinions lead to results.

And I think government support and leadership plays a very important role. I’ve been watching how the Japanese government has been funding her DX for some time now, and I find it very reassuring. Government funding has contributed greatly to the spread of deep learning in the United States as well. In the case of Japan, there are many industries that have overwhelming strengths. If I were to build something in Japan, I would choose an industry that Japan is already good at. For example, the hardware ecosystem developed in the Japanese manufacturing industry. Considering Japan’s strengths in engineering human resources and its solid industrial base, I think it is possible to bring out unique Japanese strengths that cannot be matched by other companies.

Finally, Mr. Ng and Mr. Matsuo sent messages to the viewers and Japanese business people.

Mr. Ng: The spread of electric power 100 years ago brought about a revolution in all industries, but I think AI is like a modern version of electric power. As a member of society, it is important to do our best to learn technology and think about how it will affect the people and our business. And I think it’s up to each of us to help make AI a reality, in terms of how we steer it for everyone’s benefit.

Mr. Matsuo: I think it is important to have an attitude and culture that is open to change. Technology such as deep learning is not so difficult because you just need to learn it. “AI For Everyone” can start today. I think the most important thing is to be willing to change the existing business, workflow, and corporate culture, and to adapt to the new era.

In conclusion

In this session, the importance of understanding and utilizing AI technology in promoting DX in business was discussed.

As Mr. Andrew Ng, who is also called an AI authority, said in the session that “AI is being accepted in a wider range of industries in the future,” digital shift is occurring in various fields against the background of this DX trend. .

In order to seize this opportunity and succeed with AI, many companies are demanding the development of AI human resources and new ways of doing business.