[Podcast］The surprising future of Japan’s new robot companions

Japan thinks about robotics and AI differently that the West.

In addition to their functional, productive role, a lot of thought is also given to our personal interactions, their social role, and the relationships we build with them.

Today we sit down with Shunsuke Aoki, founder of Yukai Engineering and one of the most innovative and creative thinkers on the emotional connection between humans and machines.

We talk about the future of robot companionship, how AI will change the definition of “culture”, and why the future of Japanese robotics will have a lot more participation by foreigners.

It’s a great conversation, and I think you’ll enjoy it.

Welcome to Disrupting Japan, Straight Talk from Japan’s most innovative startups and VCs.

I’m Tim Romero, and thanks for joining me.

Today, we’re going to talk about robots because I mean, hey, who doesn’t love robots?

Now, in past episodes, we’ve talked a lot about how Japan’s relationship with robots and automation is fundamentally different from what it is in the West. It’s not really about technology. I mean, technology is universal. It’s more about the personal and cultural connection to machines in general.

Well, today I have a real treat for you. We sit down and talk with Shunsuke Aoki, the founder of Yukai Engineering. Now Shunsuke may not be that well known outside Japan, but he’s one of the most innovative and creative thinkers on the topic of how humans and machines can connect on an emotional and a subconscious level.

Now, to be fair, an audio podcast can’t really do justice to Yukai engineering’s creations, but we’re going to do our best. You need to see the videos or really you need to interact with Shunsuke’s creations in person to fully understand the emotional impact.

Shunsuke and I talk about the future of robot and AI companionship, how AI will change the way we think about culture, and why the future of Japanese robotics will involve a lot more participation from foreigners.

But, you know, Shunsuke tells that story much better than I can. So, let’s get right to the interview.

Tim: So, we’re sitting here with Shunsuke Aoki, the founder of Yukai Engineering who’s creating lifestyle robots, and thanks for sitting down with us again. It’s been a while.

Shunsuke: Thank you. It’s been a while.

Tim: Yukai over the last decade and a half has been making so many cool, interesting things that just to name a few, your first big hit was the Nekomimi wearable cattier that respond to brainwaves. You have the Qoobo sort of companion pillow which has the cat tail on it. The Bocco family robot that seems to be getting a lot of traction and it’s hard to describe these on an audio podcast. But we’ll put videos and pictures on the site, but it’s hard to understand this kind of emotional impact unless you can touch and interact with these robots. With so much robotics research going on, you’ve been very focused on this sort of emotional connection. Tell me about that. Why is that important? Why do you focus on that?

Shunsuke: At the time we started a business 2011, smartphone dominated the cellular phone market right after the launch of iPhone. Japan people are skeptical and they say, a Honda are already smart and we don’t need touch screens. But three, four years later, most people accept it. As a result, I see a lot of wire connectivity, like Wi-Fi modules or Bluetooth. They’re becoming cheaper and cheaper, which makes us easy to build our own robots. I saw a possibility of personal robots that can be used as an interface not for manipulation.

Tim: Well, I understand that from sort of a technical point of view. I mean, yeah, about that time a lot of things came together to enable, like this new wave of innovation in robotics. But while most robotics startups are focused on like factory automation or increased efficiencies and things like that, you’ve taken a different path, like Bocco, for example, which has been a pretty successful product for you here in Japan. It’s a family robot.

Shunsuke: Yes. It’s designed for isolated seniors and Bocco is designed to enable family members to keep in touch easily and monitor their activity without cameras or intruding the privacy family members can leave a message to their smartphone, and the Bocco robot will speak out the message, and the seniors can make reply by recording voice message to the robot. That message is sent to the smartphone as a voice message and with a text.

Tim: So, like Bocco is a sort of cute 20 centimeter tall little robot looking thing.

Shunsuke: Right, right, right.

Tim: So why use something like that instead of using a smartphone and line chat or using Amazon Alexa or Apple Pom Pod or something like that?

Shunsuke: The idea of vocal started from my own experience as a working parent leaving my kids at home. I thought it’s not a good way to give my kid who newly entered the elementary school, so around six or seven, and he have to come home alone with the key and wait for the parents to come home for several hours. I saw smartphone is not for him because when he got a smartphone, he starts watching YouTube and I want to give him some companionship, something to keep him company.

Tim: Is there something important about having a dedicated device? So, a smartphone kind of does everything, but is there something important about having that physical device?

Shunsuke: Yeah. Smartphone can do the same job as Bocco, but Bocco is a separated physical device. When I was a kid, I was always carrying a plush toy of a bear.

Tim: The teddy bear.

Shunsuke: Yeah. Yeah. And it was my best friend, so I wanted a robot to be his best friend.

Tim: How did it work out? How did your son end up interacting with it?

Shunsuke: Interesting thing is he listens more to the robot than parents. Like when the robot told him to brush your teeth, take a bath or go to sleep, he listens to the robot very well, unlike to the parents. That was very interesting.

Tim: But for the Bocco product, rather than targeting families and young children, why did you decide to target the elderly?

Shunsuke: When we started, it wasn’t meant only for the seniors. It is useful for working parents in the age of smartphone people they’re busy with their smartphones. That was the thought.

Tim: It is interesting that people can build this emotional connection to robots. Yukai has done a lot of pilots with large companies on elder care and home care. And how have those worked out? What’s been the reaction? What’s been the uptake? How often do these seniors use it to communicate? How do they interact with it?

Shunsuke: The biggest example is Secom, the largest security company in Japan. They employed Bocco as a communication device between the isolate seniors and the operators. So, operators can chat through Bocco. It gives senior the chance to exchange messages, which is good to maintain cognitive ability for those who don’t have much chance to have conversation with people.

Tim: So, the projects have turned into long-term contracts.

Shunsuke: Right, right. It’s not a POC anymore. It actual service.

Tim: I want to get back to the comment you made just a minute ago about the importance of robot companionship. So, this is something we’ve talked about a lot in the past. You and I.

Shunsuke: Yes.

Tim: I know when you’re talking about robots as companions, you’re not talking about replacing human companionship. What kind of relationship do you think people and robots will have? What kind of companions will robots become?

Shunsuke: Robot cannot replace human or pets, pets are almost really the human. It’s almost a family member, so robots cannot replace them, but robots can be a partner. I see a lot of example in animation, of course, Doraemon. Doraemon is kind of personal coach or Nobita.

Tim: I mean, in fiction, there is that long, especially in Japan that history of robot companions. But I think even now, if you look at, it’s not robots, but its AI products that are like AI girlfriends or AI boyfriends. And if you look on Reddit and you watch how people are talking about the relationship with these AIs, the generation Z. They are building a real relationship, a real emotional connection. They know it’s not human. They’re not like delusional. But both men and women seem to be forming this kind of unique emotional bond.

Shunsuke: Maybe, but it’s not something that substitutes humans.

Tim: No, I don’t think so. I don’t think that most of the people using these Apps view it as a substitute. They view it as something different and unique. Sort of like the Doraemon robot coach. With all the rapid development in AI over the last couple of years, especially with large language models, being able to enable computers to communicate in a much more human-like way. What’s the impact you think that AI is going to have on robotics, particularly these emotional, these lifestyle robots?

Shunsuke: I think we have a huge possibility we can finally give robots conversational feature. And again, I think it’s not to substitute humans. Because people make friends and people are very sensitive about subtle difference between humans and…

Tim: Well, that’s true. But I guess I think it is possible to develop a healthy emotional connection to robots or AI or whatever. I’m not sure exactly what that looks like, but I think it’s possible. But I guess it’s also possible to develop unhealthy connections and relationships. And what do you think about the danger of — because it’s not a substitute for human relationships, but human relationships are hard, they’re troublesome.

Shunsuke: Yes. Yes.

Tim: So, do you think there’s a danger of these lifestyle robots, especially when coupled with modern AI to start replacing these kind of relationships? Not intentionally, but unintentionally, just because it’s so much simpler than maintaining human relationships.

Shunsuke: We hope that the AI can be a personal health coach or personal language tutor, or personal financial advisor to not only give you advice, but helps you to keep your motivation and help you make a healthy habit. But yes, it can be used to make you build unhealthy habits.

Tim: So, you think it’s really the responsibility of the AI developers and the robot makers to ensure that people don’t form these unhealthy attachments to steer people away from those kind of unhealthy interactions?

Shunsuke: Right. We already have the kind of problem with TikTok whose AI recommend you to engage more and more to their platform, which is very toxic. Also, YouTube does similar things.

Tim: Well, I agree. That’s a real danger. But I think a lot of what you’re developing these kind of lifestyle robots, they’re not ad supported. You’re not selling advertising, so you don’t make more money the more people engage with it. So, lifestyle, robotics, the emotional connection, I guess if it is paired up with an advertising model, that becomes very dangerous.

Shunsuke: Yeah. AI have an ability to guide people to take action, to buy some products or maybe some religions.

Tim: Yeah. Well, I think we’re starting to see that the AI use in social media now, a lot of it’s being used in advertising and political campaigns.

Shunsuke: Yes. Yes. Right, right, right.

Tim: Getting back to robotics. So, Japan has always been incredibly strong in robotics. I’ve always been quite skeptical of human-like robots. I mean, it’s interesting from an engineering point of view, but it seems like nobody really wants them. Boston Dynamics in the US, the Atlas is amazing. They can’t sell it. It doesn’t do anything. And like Azimo for a long time, SoftBank’s pepper. These kind of robots just don’t succeed in the marketplace. Why is that?

Shunsuke: That’s an interesting question. And we see many AI companies in US also started to create humanoids, and they say the robots will help their manufacturing process. I’m very skeptical about the idea of humanoids robots can be used to help manufacturing.

Tim: But we’ve been seeing this for 40 years, 30 years. So, do you think there will be a place for these humanoid robots?

Shunsuke: I don’t see any feasibility. But it’s always an interesting subject for engineers and also it’s I think it’s deeply rooted in human nature to create something similar to us.

Tim: Oh, okay. Yes. That’s definitely true. Yeah. I think you’re right. That’s going back from the dawn of civilization. Maybe we just can’t help it. But the most effective robots, like in industrial settings, it’s not trying to make something that looks like the worker, it is like looking at the task and trying to create the most efficient way of doing it, like extracting that useful industrial motion out of the process and building a machine. Or I guess kind of you guys are doing the same thing, with like the Qoobo, the cat robot. It’s kind of extracting the emotional component. It doesn’t look like a cat. I mean, maybe a little bit, but not really but it gives you that emotional connection.

Shunsuke: Right. It doesn’t have to look like a cat.

Tim: Right, right. Well, I guess compared to like Sony’s Aibo, which looks like a dog and behaves like a dog, but interact with it feels nothing like interacting with a dog. Whereas the Qoobo looks nothing like a cat, but somehow interacting with it feels like a cat.

Shunsuke: Yeah. That’s right.

Tim: Let’s talk about Yukai’s business strategy and growth strategy, because as interesting as consumer robotics are in Japan, and as important as I think these lifestyle robots, these emotional evocative robots are, across the board they’re having a hard time making money. So, like the Lovot robot couldn’t survive. Aibo has, I don’t think they’ve ever really made money on it. It’s Sony, so they can keep selling it, but I don’t think they make money on that. Pepper didn’t succeed. And actually now that I think about it, even tools like Alexa and Google Assistant never really took off like they expected. People use them to like set timers and play music, but not as this kind of interactive assistant companion. So, what’s the plan for succeeding in this market where we’re so many of these lifestyle robotics, lifestyle AIs have failed?

Shunsuke: We are committed to the lifestyle robotics market. Many people needs empowerment to lift their potential.

Tim: So, do you think it’s a matter of timing? Do you think that, for example, with the additional stress that not just young people, but all of us feel from like social media. Do you think it’s just a matter of timing before these emotional robots, the lifestyle robots, emotional support robots, succeed in the market? Is it just a matter of waiting for that demand to get high enough?

Shunsuke: I believe we still lack some technology. Generative AI could be the missing technology.

Tim: Is that something you guys are working on now?

Shunsuke: Yes, of course. We’ve added several features to Bocco that uses generative AI to help children to develop their daily habits.

(Continued on Disrupting Japan)

Click here for the Japanese version of the article