Realistic digital humans interact with audiences in real time. Autonomous systems streamline complex logistics. And AI-powered language tools break down communication barriers on the fly.
It’s not science fiction. This is the startup scene in Tokyo.
Supercharged by AI – and world-class academic and industrial power – the region has become a global innovation hub. And the Creation of NVIDIA the program is right in the middle.
With more than 370 AI-based startups enrolled in the program and a 250,000-strong NVIDIA developer community, Japan’s AI startup ecosystem is as bold as it is rapidly evolving.
This week’s NVIDIA AI Summit Japan puts these achievements in the spotlight, capturing the region’s relentless innovation drive.
NVIDIA Founder and CEO Jensen Huang and SoftBank Group Chairman and CEO Masayoshi Son opened the summit with a fireside chat to discuss the transformative role of AI, with Jensen diving into the ecosystem growth of AI in Japan and its efforts towards sovereign AI.
Sessions followed with leaders from METI (Japan’s Ministry of Economy, Trade and Industry), the University of Tokyo and other key stakeholders. Their success is no accident.
Tokyo’s academic powerhouses, global tech and industry giants, and its tech-savvy population of 14 million provide the foundations of a global AI hub that spans the bustling scene from startups in Shibuya to new hotbeds of tech development in Chiyoda and beyond.
Energizing the Japanese creative class
Iconic works, from anime to manga, have not only redefined entertainment in Japan: they have etched themselves into global culture, inspiring fans across continents, languages and generations.
Today, Japan’s vibrant visual pop culture is spilling over into AI, finding new ways to surprise and connect with audiences.
Take Sali, the digital celebrity of startup AiHUB.
Sali is not just a character in the traditional sense. He is a present, responsive and realistic digital being. She blinks, she smiles, she reacts.
Here, AI is doing something revolutionary by going under the radar to redefine the way people interact with media.
At AI Summit Japan, AiHUB revealed that it will adopt the NVIDIA Avatar Cloud Engineor ACE, in the lip-sync module of his digital human frame, providing Sali with nuanced expressions and human-like emotional depth.
ACE not only makes Sali relatable, it places her in a league of characters that transcend screens and pages.
This integration reduced future development and management costs by approximately 50% while improving the expressiveness of avatars, according to AiHUB.
SDK adoption: from hesitation to high velocity
In the global race for technology, success doesn’t always depend on the heroes we expect.
The unsung stars here are SDKs – those sets of tools, libraries, and documentation that take the guesswork out of innovation. And in Japan’s rapidly evolving AI ecosystem, these once-neglected SDKs are sparking an unlikely revolution.
For years, Japanese tech companies have treated SDKs with caution. Today, however, with AI advancing at the speed of light and NVIDIA GPUs powering the engine, SDKs have moved from a quiet corner to center stage.
Take NVIDIA NeMoa platform for creating large language models, or LLMs. This is quickly becoming the backdrop for Japan’s latest wave of AI-based real-time communications technology.
One company at the forefront is Kotoba Technologies, which has cracked the code of real-time speech recognition using NeMo’s powerful tools.
Thanks to a key grant from the Japanese government, Kotoba’s language tools don’t just capture sound: they translate it live. It’s a blend of computing power and human ingenuity, which is redefining the way multilingual communication happens in non-English speaking countries like Japan.
Kotoba’s tools are used in customer call centers and for automatic creation of meeting minutes in various industries. It was also used to do live transcription during the AI Summit Japan fireside chat between Huang and Son.
What if LLMs are the engines of Japanese AIthen companies like APTO provide the fuel. With NVIDIA NeMo Curator, APTO is a game changer in data annotation, handling the intensive prep work that makes LLMs effective.
By refining data quality for large clients like RIKEN, Ricoh and ORIX, APTO has mastered the art of filtering valuable signals from noise. With tools like WordCountFilter – an ingenious mechanism that eliminates short or unnatural sentences – performance is optimized.
APTO’s data quality control improved model accuracy scores and reduced training time.
Developers across Japan are looking to quickly move toward AI and are adopting SDKs to go further, faster.
The power of cross-sector synergy
The gears of Japan’s AI ecosystem are turning increasingly in sync thanks to NVIDIA-powered infrastructure that allows startups to build on each other’s advances.
As Japan’s population ages, solutions like these address security needs as well as a growing labor shortage. Here, Ugo and Asilla took on the challenge by using autonomous security systems to manage facilities across the country.
Asilla’s cutting-edge anomaly detection was developed with security in mind, but is now finding applications in healthcare and retail. Built on the NVIDIA DeepStream And Triton Inference Server SDK, Asilla’s technology doesn’t just identify risks, it responds to them.
In high-stakes environments, ugo and Asilla’s systems, powered by NVIDIA Jetson platform, are already in action, identifying potential security threats and triggering responses in real time.
NVIDIA’s infrastructure is also at the heart of Kotoba Technologies’ language tools, as well as AiHUB’s realistic digital avatars. Operating on an AI basis, these different tools seamlessly connect media, communication and human interaction.
The story behind the story: Tokyo IPC and Osaka Innovation Hub
All of these startups are part of a larger ecosystem that is accelerating Japan’s rise as an AI powerhouse.
Leading the charge is UTokyo IPC, the wholly-owned venture capital arm of the University of Tokyo, which operates through its flagship accelerator program, 1stRound.
Co-organized by 18 universities and four national research institutes, this program serves as a link between academia and industry, providing practical advice, resources and strategic support.
By championing the real-world deployment of early-stage deep technology innovations, UTokyo IPC is igniting Japan’s academic innovation landscape and setting the standard for others to follow.
Meanwhile, the Osaka Innovation Hub, OIH, is extending this momentum beyond Tokyo, providing startups with coworking spaces and networking events. Its startup acceleration program allows early-stage projects to be brought to market more quickly.
Rapidly evolving centers like these are at the heart of Japan’s AI ecosystem, providing startups with the mentorship, funding and resources they need to move from prototype to fully commercialized product.
And thanks to NVIDIA’s accelerated computing technologies and the Inception program, Japan’s fast-moving startups are joining forces with AI innovators around the world.
Image credit: ugo.
