News Release
2026.07.16
avatarin Pioneers Sovereign AI Agent Foundation in Japan with
Operational NVIDIA B300
TOKYO, Japan– July 16 2026 –
avatarin Inc., a Tokyo-based AI and robotics startup, today announced that it is advancing a sovereign AI agent foundation in Japan, starting with an NVIDIA B300 system that is currently operational.
Empowering Enterprise Control Over Sovereign AI
The initiative is designed to help enterprises deploy AI agents across digital and physical customer interfaces while maintaining control over their data, operational knowledge, brand intelligence, and AI improvement.
As generative AI and AI agents move from experimentation to enterprise deployment, companies increasingly need to determine where sensitive data is processed, how operational knowledge is managed, and how AI systems are continuously improved. This is particularly important for enterprises in regulated or mission-critical industries in which customer interaction data, speech data, operational manuals, and voice-of-customer insights are valuable core assets that must be protected.
Securing Insights from Sensitive Customer Data
A key focus of the initiative is enabling enterprises to derive deeper insights from sensitive customer interaction data, including voice-of-customer signals and speech data, without unnecessarily moving proprietary data outside controlled environments. By combining operational NVIDIA Blackwell Ultra-accelerated computing with enterprise-controlled data pipelines, avatarin aims to help organizations analyze speech and customer interaction data more securely, uncover deeper business insights, and continuously improve AI agent response quality, domain understanding, and brand alignment.
The Vision: “One Intelligence, One Brand, Every Interface”
avatarin is building toward its “One Intelligence, One Brand, Every Interface” vision. The company aims to enable enterprises to deliver a seamless, unified AI agent experience across an ecosystem of touchpoints—including contact centers, web applications, mobile apps, physical retail stores, airports, robotics, and digital signage.
Starting with its operational NVIDIA B300 system, avatarin will continue exploring strategic opportunities to leverage NVIDIA technologies including NVIDIA Nemotron open models for agentic AI and real-time speech AI, NVIDIA Audio2Face for audio-driven facial animation and lip-syncing.
Executive Commentary
“AI agents must evolve beyond mere conversational interfaces,” said Akira Fukabori, CEO of avatarin. “Their true value lies in their ability to seamlessly synthesize an enterprise’s unique data, operational knowledge, brand intelligence, and voice-of-customer signals into a cohesive experience across every single touchpoint. With our operational NVIDIA B300 system as a foundation, avatarin aims to help Japanese enterprises analyze sensitive customer interaction data more securely, uncover deeper insights, and command full ownership over their AI future.”
Related Announcement
For more details, please see the announcement here.
■https://nvidianews.nvidia.com/news/japans-enterprises-and-startups-build-industry-specialized-ai-with-nvidia-nemotron-open-models.
About avatarin, Inc.
avatarin is a Tokyo-based AI and robotics startup building the foundational nervous system for enterprise intelligence. The company empowers organizations to deploy secure, sovereign AI agents capable of delivering “One Brand across Every Interface.”
■Learn more at https://about.avatarin.com/en/.
Media Contact
■Company: avatarin Inc.
■Department: Public Relations
■Email: pr@avatarin.com
Contact Us