Finance TnT Executive Editor Loren Moss was able to catch up with Kukita Shinya (above left), the globetrotting executive engineer for the technology giant NEC. They were able to talk in the Caribbean port city of Cartagena during the annual ANDICOM ICT industry event. Robotics and Artificial Intelligence (AI) are irrevocably changing the relationship between man & machine. In this conversation, Shinya talks about what we can do to prepare, and what to expect in the coming world.
Finance TnT: Japan is famous for now having an aging population and its interesting because Japan is also a leader in consumer robotics. NEC used to be a consumer company, now its more of an industrial and government company, but we see some of your colleagues that you left behind in the consumer space coming out with different products and one of the things that people are saying is that there are ways now with an aging population with a low birth rate, we are going to see more and more technology used to accompany and to help take care of family members and things like that. That´s an interesting thing, I think that brings up some unforeseen interesting applications. How do companies and how do societies deal with the change? Where technology before was: maybe I am going to use a vacuum cleaner, or watching TV. But now how do societies have to adapt to understand this new role and not even just at the consumer level but the situation with security? So, people are worried about privacy and people are worried about facial recognition, and yes, it is for my security, but what about my privacy? What are the issues that you see in maybe governments having to grapple with, as NEC is a leader on this field?
Kukita Shinya: Okay, you´ve touched on many different things. Exactly like you said, AI (Artificial Intelligence) has basically very high expectations but at the same time, there are some concerns. The concerns are divided into two primary aspects, one of them is the privacy issues, the other one is job security. I´d like to talk about both. However, before we start, regarding AI, NEC has been working on Artificial Intelligence for 50 years, that means AI is at the peak of its height at this time but its not the first time, it’s the third time, so the result in AI becomes boom! This intelligent automation is characterized by deep learning and machine learning, and this deep learning and machine learning only becomes possible with the very affordable cost of Big Data and also computing. So, the concept itself is there and has been there for more than 30 or 40 years, but it has never been realistic to access that Big Data and have the processing power to consume that much. So, once again it becomes only available in the past few years, so those are the AI conditions. NEC has been doing this AI thing, starting from the OCR and automated dictation, those were the first applications of the AI per se, but nowadays, nobody calls that AI any more.
Finance TnT: You are right! I remember when that first came out I was like: WOW! I remember when Dragon Dictate came on the market in the 1990´s, packaged voice to speech dictation software you could buy, and it worked surprisingly well for that time. Today it wouldn’t be acceptable but at that time it was pretty good.
Kukita Shinya: And at that time people said it’s the greatest and latest. But at the time people would say that´s a piece of software, its not AI, and that happens so many times. Regarding AI, at this point, there are two different stages or types of products. One of them is like the face recognition and the video analysis, those types of AI were trained by NEC in-house before we launched the software, so that is simple to use for the final user, like packaged software. The other type of the software is the deep learning type software engines, and for those we need to have a learning period, we need to prepare the big amounts of data and process it, so these operations take more time and more expertise.
At this point, that type of AI is only applicable for big projects, typically big enterprises or governments who have the access to the big amounts of data. So, we have two completely different types of AI at this point, video surveillances, face recognition, those are the types of AI that we have available for almost anybody. The application we have in the Colombian football stadium is another example of those type of things. With the other type of AI it’s not that simple. We need to have preparation and heavy work to make it happen. So, that´s the AI situation. Regarding the issue of privacy versus security? So we notice there are two completely opposite opinions, optimistic and pessimistic. Optimists say it happens, not only for AI, but every time new technologies are introduced, when the new model is introduced, the other one is gone. So, its not special for AI, that´s one way of seeing it. The other way of seeing it, AI is so universal, it will affect so many industries at the same time, it will have a big impact towards job security, not necessarily to reduce the jobs as a total, but the activity or the expertise required for human workers will be different from the past.
Finance TnT: Honda´s US headquarters is outside of my hometown in Columbus Ohio. When I was maybe around high school age a lot of the kids said: “we are going to go and try to work at Honda.” Some of us went to college, others went to Honda, but its interesting because now you go into an auto plant and its almost all automated, its like people can go home and the car will still get assembled, but someone must make the robots and program them, and even though lately some say robots can make robots, some one has to still program them, and even as programming evolves and changes, creativity, management and judgement are still there.
That bring up an interesting question. In the US, you hear about the political cycles—and maybe in Japan too—but a lot of politicians say: We have to bring back many manufacturing jobs! I say wait a minute, you want a good job, or you want a manufacturing job? My friends who worked in the factory liked to have a pay check, but they didn’t really like the work. They thought the work was boring and awful and even though they had a nice boss and it was a nice company, it wasn’t really the best of the jobs. I couldn’t take it, I would go crazy.
How do we prepare the next generation? I think its complicated because there are always government policies, political interests, and different things like that. What should the next generation study? Not only if you are a professional and you want to go into IT and software, if you want to be in IT study this, if you want to be in AI, study that. On the other hand, like I said, some of then went to college, some of them went to the factory, for those that maybe colleges are not for them, or IT is not for them, what is their place? Where do they fit in this new society that´s coming out, shaped by all these different disruptions?
Kukita Shinya: Once again, you touch on many different things. The first thing is, acceptance by the society is based on the maturity level of the society. I´m not necessarily saying an advanced society is better! One particular thing about Japan is that it is one of the fastest aging countries, Singapore is the same way. We will not have enough work force, not necessarily in industry, but also in other areas as well. So, we have no choice other than introducing automated robots and AI to help us maintain our own lives, so that´s one situation.
In the US and in many other countries that are still growing, they have the younger generations and have different perceptions about the introduction of AI. I don’t think there´s one single answer about what needs to be done, or about how we should prepare, once again it depends on the maturity level of the society. The other thing you mentioned about AI, the robots in the workplace or manufacturing facilities is inevitable, but one interesting thing I have noticed is that the robot in the manufacturing plant is not sophisticated enough to do certain duties, they don’t have enough vision.
Finance TnT: Situational awareness.
Kukita Shinya: Yes, situation awareness is not enough. They could swing their arms and hit people, so, that´s the reason why we have a fence to isolate the human work force from the robot force. Having robots at home is a much more challenging situation. So, until we feel comfortable in that aspect, we will not have the robots next to you. The penetration of the robots will start in those isolated environments and then based on the maturity of the awareness capability of the robot we will decide the location where we can apply the robot in a house.
Finance TnT: It seems like up until now, robots have excelled on doing repetitive tasks: Put the screw in the hole, but now you look at some of the things that are coming out of Artificial Intelligence, Right now we don’t really see them in robots but we see them in smart appliances like Amazon has Alexa, and Apple has Siri and all these things, you can kind of have a conversation with them, and its almost like we have this world of software and these appliances that are interactive but not in a connected, full motion way and then we have these robots that are connected but are not yet interactive.
I think the fascinating thing that we are going to see in the near future is when these two things merge together. So, when you have a robot or something that has the ability to move around, but it has the Artificial Intelligence that you see on some of these devices like in Alexa, then I think this takes us into a whole different world. My grandmother had Alzheimer and a lot of times specially for my mother her conversation was very frustrating because my mom wanted to talk to her mother, but her mother wasn’t really in there, I hate to say it like that, its very frustrating but you have to talk to them to keep their minds stimulated.
Not in every case, but sometimes you can use Artificial Intelligence and use things like IoT (Internet of Things) for example to track my grandmother and detect when she leaves the house and wanders off. Nobody would know where she was, she thought she was in a different city, and how you can use things like IoT not only to track the person but maybe some kind of Artificial Intelligence that´s interactive, that´s talking to the person and saying: Maybe you want to come back, or maybe you want to sit in this bench, I think that there is all these different ways. I have seen the manufacturers have things like electronic pets, what if you had a patient that had Alzheimer or not even Alzheimer, for example autism, were there are some kinds of disabilities where the person maybe didn’t have the decision-making skill. So, by having that kind of companion, that if the person leaves the house, the robot follows the person and if they need help it says: Hey, you are outside of the perimeter…Please follow me.
I think there are so many applications. What´s the role of companies like NEC? What is the role of the companies that are making these types of things? What leadership can companies take in shaping the future vs waiting for the consumer, waiting for society, waiting for the government. How do companies help shape the direction where we are going with this?
Kukita Shinya: Before answering the last question, you brought up three main points, let me talk about the first point: The connected things and the interactive part, the AI or the robot, that means that from the past until today the majority of the job security things are related to blue collar activities, but now the integrated robot is doing paralegal activities or the countdown has started where the robots are coming for the white-collar jobs,so that´s one aspect.
The second point is, I would say Japan or the Japanese people are one of the most robot-friendly countries, maybe it’s because of the cartoons and animations, vs the Americans who learn everything about robots through Hollywood movies, where the robots try to dominate the people. So, that might be the difference, based on that familiarity with the robots, the pet robot to talk with elderly people is a very common practice in Japan, and then there are other alternatives for Alzheimer and those cases. Not necessarily to cure diseases, but at least to lighten the burden for the people next to the patients. So, for example in the case of your mother, who had a hard time, some part of that can be eased by the robot.
What is NEC doing in that aspect, NEC created a group of people that we invited from different industries, we have very different people, diverse expertise and views of society; a group of people that are talking about what´s going to be in ten years, twenty years, fifty years later. So, we always put the human in the center, so, everything needs to be for humans, whenever we talk about Artificial Intelligence or Automated Security, everything is at the end for humans.
Finance TnT: So, its technology serving humans not vice versa.
Kukita Shinya: That´s right.