How does an android think? How do they feel? Finding an answer to these questions is specifically a matter of the now that we are currently experiencing. It’s obvious that AI and other advanced technologies are already part of the human experience, and finding the right approach to relate to them is the key to all this. Keiichiro Shibuya is releasing today one of his greatest creations, Android Opera Mirror. To delve into it is to start to understand how there are magnificent and heartfelt chances of getting closer to technology in order to relate to it. So, the answer? We find it within Keiichiro’s artistry.
He first touched a piano at the age of six, and still today believes it is the most advanced technology in music. Keiichiro Shibuya is a Japanese composer, musician, and artist based between Tokyo and Paris, and he graduated from the Tokyo University of the Arts’ Department of Composition. His work encompasses cutting-edge electronic music, piano solos, operas, film scores, and sound installations. Through his music compositions and collaboration with artists and scientists, he challenges the boundaries between humans and technology, and life and death. In 2002, he founded the music label ATAK, working across live performances and recorded music.
Shibuya received the 30th Japan Movie Critics Award for the film Midnight Swan. He has been recording cutting-edge electronic music since the early 2000s, and has also worked with high-profile brands like Gucci, Prada, Nike and Lexus. Keiichiro was appointed as a visiting professor at Osaka University of Arts and established the Android and Music Science Laboratory (AMSL). The Lab was designed by architect Kazuyo Sejima (SANAA), and it fosters new ways of creating music with androids and artificial intelligence. Shibuya’s previous work includes the innovative vocaloid opera The End, starring Hatsune Miku in 2012, Scary Beauty in 2018, and Super Angels in 2021. He also created the modern production Parade for The End of The World in 2016, that recreated the collaborations of Satie, Picasso and Cocteau. Android Opera Mirror is one of his biggest projects yet; it was first shown for Expo 2020 Dubai, and a development of it was premiered at the Théâtre du Châtelet in Paris, in 2023.
Rooted in the theories he investigated with his acclaimed android opera stage productions, Android Opera Mirror is an ambitious, multifaceted sci-fi epic that wonders how an AI-equipped humanoid robot might react to the most human artistic experiences, conducting an orchestra and performing as a solo vocalist. The only human element on the album is Shibuya’s tender piano playing; even the orchestral elements, which were tackled by humans in the stage show, are built from artificially generated data that’s then mixed and processed like a normal orchestral production. It’s Shibuya’s opportunity to present a scenario where human beings and android-like technology are not in opposition to each other, but in harmony.
Set some time in the far future, the piece weaves together memorable, iconic themes, experimental electronic motifs and bombastic neoclassical flourishes, posing questions that touch on philosophy, religion, literature, art and technology. The concept has been in development since 2017, and Shibuya has already staged variations in Australia, Japan, Germany, the Middle East and Paris, but this piece isn’t just an archive of these original performances – it’s a fresh examination of the source material that continues to blur the boundary between man and machine. Standout tracks such a Mirror, Scary Beauty or Borderline lead us to a critical but possibly hopeful outcome: “the world you have once loved is no longer yours.”
Midnight Swan (Android Opera ver.) fleshes out the popular original theme with lush orchestrations and GPT-generated android vocals. The track is the album’s unofficial centrepiece, adding a glittering chamber pop twist to a memorable melody that’s already racked up close to seven million views on YouTube, and on Lust, Shibuya introduces the esoteric Buddhist teachings of the Rishu-kyō.
His work is here to challenge how people relate to technology, not only for its countless possibilities but because it’s probably more than ever that we need to learn different ways to engage with it. And Shibuya has both success when creating such a vast project as well as handling emotions as the core of the opera, as a testament to the present and as desire for the future.
We had the chance to talk to the artist himself to learn more about his work and his career in this interview.
Hi Keiichiro, how are you? How are you feeling these days as we’re getting closer to the release of Android Opera Mirror?
I'm both excited and a little nervous. I'm looking forward to seeing how people interpret the concept and music of this album.
Congratulations, it is a spectacular piece, there are so many layers of work within it. I wonder when you first thought about making this project, and what were the first steps to approach such complex work?
Thank you. Android Opera was actually a spontaneous idea that came to me in 2013, right after I had premiered The End, a vocaloid opera starring Hatsune Miku that explored themes of life and death, in Paris. At the time, the director of Théâtre du Châtelet asked me, “Keiichiro, what are you going to create next?” and I instinctively responded with this concept. I never imagined it would grow into such a large-scale project.
The first creative step was to find a compelling text that would be interesting when sung by an android rather than a human singer. I spent a lot of time thinking about what kind of poetry could create a meaningful and intriguing sense of dissonance.
The first creative step was to find a compelling text that would be interesting when sung by an android rather than a human singer. I spent a lot of time thinking about what kind of poetry could create a meaningful and intriguing sense of dissonance.
The role of AI was key to the development of this piece, but more specifically to the character presented to us as an android. What was the process of creation like? I wonder if any error or unexpected results in the realisation of this piece have remained part of the nature of the android character.
This android integrates multiple layers of AI and various programmes. For instance, we often think we are in control of technology, but in a way, we are also its slaves. When a technological error causes the android’s conducting to spiral out of control, the human orchestra must still follow its lead. This serves as a metaphor for our contemporary relationship with technology.
Additionally, the android can react to human piano and vocal performances, improvising melodies in response. The lyrics generated by GPT evoke an image that is inconceivable in a human-centric world — one that celebrates the end of the world. In a reality where the apocalypse feels inevitable, is there any perspective beyond mere despair? This question drives the opera, unfolding in various forms. And at its centre stands an android as the lead singer — an unconventional and unprecedented character in the realm of opera.
Additionally, the android can react to human piano and vocal performances, improvising melodies in response. The lyrics generated by GPT evoke an image that is inconceivable in a human-centric world — one that celebrates the end of the world. In a reality where the apocalypse feels inevitable, is there any perspective beyond mere despair? This question drives the opera, unfolding in various forms. And at its centre stands an android as the lead singer — an unconventional and unprecedented character in the realm of opera.
I really like the voice of the android. It's very interesting to listen to what they sing, because it doesn't seem superficial, cold or distant; on the contrary, it's warm, it evokes tenderness, and it shows the possibility of future and advanced technology as an ally of the human being. Was that important to you, given that you are the head conductor of the work?
Indeed. The voice was one of the most important elements. The idea that synthetic sound should be cold simply because it’s artificial is tied to an outdated perception of technology. While the android’s voice isn’t human and is created from electronic sounds, I wanted to shape something that still carried warmth — a sense of romanticism and sentimentality. I believe that encountering an entirely new kind of being and feeling moved by it is an experience that can be created through voice and music.
Based on a solo piano composition you wrote for Eiji Uchida's film, Midnight Swan is an incredible and beautiful piece. You took it to another level when you did the version for Android Opera Mirror. Here, it becomes a fantastic pop song with depth and more colour. Is it difficult to rework such a specific song for a project like this?
Creating different variations of a single piece comes naturally to me, so it wasn’t difficult. I actually enjoyed the process, taking the romantic and emotional essence of the original piano piece and elevating it into something celebratory, while also embracing the virtuality that comes with an android’s voice.
You also brought and upgraded I Come From The Moon, originally composed for a Gucci campaign. How different is your work when you are making the soundtrack for a fashion project?
Whether it’s for fashion or any other visual element, I focus on understanding the concept and intent behind it. Once I grasp what kind of music would complement it, the rest of the process isn’t all that different. When working on my own projects, I compose based on my own vision and concept, but the fundamental approach doesn’t change much. It’s like wearing a jacket to a restaurant versus deciding whether or not to wear one to a café — the difference is there, but it’s natural to me.
Do you remember when the first spark between you and technology happened in your life? I find it interesting how it is an instrument for you, and how that affects how you may have a framework of thinking, but also, given how quickly it develops, how has that relationship evolved through the years, from the type of computers we were using back in 2000 to the use of AI in recent years?
It started when I first touched the piano at the age of six. Even now, I believe the piano is the most advanced technology in music.
Another key moment was in the late 1990s when I began making music on a Mac PowerBook G3. Suddenly, you didn’t need music theory, synthesisers, or electronic instruments, just a computer was enough. That was the biggest revolution in music since the piano. I embraced a minimalistic production style, focusing on minimal electronic music and noise for several years in the early days of my label, ATAK. It was actually a great opportunity to strip away my classical education and start fresh.
Over the past 15 years, I’ve been working to merge classical methods with electronic music. And now, AI has emerged, it feels like the bridge between the two, which is why I find it so exciting to work with.
Another key moment was in the late 1990s when I began making music on a Mac PowerBook G3. Suddenly, you didn’t need music theory, synthesisers, or electronic instruments, just a computer was enough. That was the biggest revolution in music since the piano. I embraced a minimalistic production style, focusing on minimal electronic music and noise for several years in the early days of my label, ATAK. It was actually a great opportunity to strip away my classical education and start fresh.
Over the past 15 years, I’ve been working to merge classical methods with electronic music. And now, AI has emerged, it feels like the bridge between the two, which is why I find it so exciting to work with.
The opera and its different stages have been presented in recent years in live shows in Australia, Japan, Germany, the Middle East and Paris. What was it like to put together the performance? How did you work on the visuals, and the performative implementation of the physical android scene on set?
It was a collaborative process, while communicating my vision to programmers, visual artists, and lighting designers, I also listened to their input and built the show together with them. That’s why the performances were always a work in progress, constantly evolving. When we staged it in Tokyo last June, I finally felt that it had reached a certain level of completion, which made me decide to release the album.
Beyond the controversy it generated when it was published, you have been inspired by Michel Houellebecq's The Possibility of the Island. I see that in both his novel and your opera there is a message that stands out above the plot: the possibility of a future in which technology and human beings come together. What inspired you from this book? And are there other cultural pieces that were an influence during the making of Android Opera Mirror?
The Possibility of an Island shifts between two different timelines, and within it, there is a very direct love letter. While it’s a story about human relationships, the most deeply human poetry in the book stands out, and I thought it would be interesting to have an android sing it, creating an unexpected contrast.
I also incorporated the final works of human novelists and philosophers into the libretto. The Decay of the Angel was Yukio Mishima’s last novel, and On Certainty was Ludwig Wittgenstein’s final work. The maturity and instability present in a person’s last creation are things an AI can never experience, which is why I found it compelling to have an android sing them.
I also incorporated the final works of human novelists and philosophers into the libretto. The Decay of the Angel was Yukio Mishima’s last novel, and On Certainty was Ludwig Wittgenstein’s final work. The maturity and instability present in a person’s last creation are things an AI can never experience, which is why I found it compelling to have an android sing them.
Did your previous work with Hatsune Miku help when creating this opera?
I believe so. In the sense of creating an opera that isn’t centred on human performers, which challenges the human-centric perspective often seen in European opera, there is a clear connection. Android Opera represents the maturation of that concept and its evolution into a new form of expression.
“The world you once loved is no longer yours” is a very relevant statement we can hear in Borderline. I know it's a very broad question, but how do you see the world today and the different uses people make of technology? Unfortunately, it is not always for a good purpose.
This lyric was actually generated by an early version of GPT, before the birth of ChatGPT. At that time, GPT wasn’t as advanced or as flexible as today. Yet, it still produced such a striking and precise text, which continues to amaze me. Even without highly sophisticated technology, the way humans interpret what is generated can make something deeply thought-provoking or emotionally resonant. Rather than focusing on how humans use technology, I think it’s more important to consider how we interpret the messages that technology produces.
You set up the Android and Music Science Lab (AMSL) at Osaka University of the Arts some years ago, can you tell us a bit about how it came about and how it is used today?
When Android Alter 4 was created, I wanted to establish a lab that connects androids, music, and technology, bringing together engineering, AI, and musical expression. The lab’s interior design was created by Kazuyo Sejima, the world-renowned architect from SANAA. Artists and programmers come here for residency-style collaborations, where they work intensively on productions and experiments. A bionic artist Viktoria Modesta even came from LA to film a project with us.
After all these years with your particular approach to music creation using technology, what is the main reflection you have learned?
To embrace the difficulties and complexities of collaborating with scientists rather than resisting them. That misusing technology, and even its errors, often leads to the most exciting results. And most importantly, to avoid trying to use it correctly.
Finally, what's next for Android Opera Mirror, and for your career?
Android Opera will soon undergo another major evolution, which I hope to announce in the near future. At the same time, I’m also working on an entirely new project, aiming for a 2026 or 2027 premiere.
As a composer, I believe that theatre music, film scores, and religious music will continue to exist throughout my lifetime. I’ve worked in these fields before and want to keep refining my approach. At the same time, because these forms have such strong, established structures, they are well-suited to integrating new technologies.
Thank you very much for taking the time, Keiichiro. Congratulations on this masterpiece, and all the best for a bright future!
Thank you for your insightful questions.