NVIDIA Invests in Taiwanese AI Digital Twin Startup Metai

Nvidia Invests in Taiwanese Digital Twin Startup MetAI

Nvidia is significantly expanding its focus on robotics and industrial AI applications. This is evidenced by the launch of its Omniverse platform and, more recently, Mega, an Omniverse Blueprint framework designed for creating digital twins. The company is also actively investing in startups specializing in digital twin technology to accelerate this initiative.

Rapid Digital Twin Creation with AI

MetAI, a company based in Taiwan, has pioneered a model capable of rapidly generating “SimReady” digital twins. Utilizing AI and 3D technology, the system converts CAD files into fully functional 3D environments in a matter of minutes.

This innovative approach has attracted Nvidia, who is leading MetAI’s first funding round. The $4 million seed round marks Nvidia’s inaugural investment in a Taiwanese startup.

Investors Join the Round

Alongside Nvidia, the funding round includes participation from a diverse group of strategic and financial investors. These include Kenmec Mechanical Engineering, Solomon Technology, SparkLabs Taiwan, Addin Ventures, and Upstream Ventures.

Generative Physical AI and the Role of Digital Twins

The emerging field of generative physical AI depends on highly accurate simulated environments. These environments are crucial for training and validating robots used in autonomous systems, allowing for the development of operational AI before real-world deployment. MetAI believes its digital twins will be central to this process.

Addressing the Barriers to Digital Twin Adoption

According to Daniel Yu, CEO and co-founder of MetAI, “Digital twins have long been seen as a barrier to entry for physical AI due to the months or even years of effort required for development.”

MetAI specializes in AI-powered digital twins specifically designed for advanced semiconductor fabrication plants (fabs), intelligent warehouses, and automation solutions. The company also generates synthetic data within these AI-enabled digital twin environments.

The Founding of MetAI

Renton Hsu, MetAI’s co-founder and CTO, brings expertise in both 3D engineering and AI. He initially encountered the power of digital twins while developing enterprise AI software. They served as a practical solution when clients lacked sufficient data for system training.

Recognizing the potential, he applied this concept to 3D systems, integrating 3D technology with AI to create synthetic AI and 3D solutions. He then collaborated with Yu and a third co-founder, Dave Liu (COO), to establish MetAI.

Recognition from Nvidia

MetAI’s innovative approach earned them first place in a competition hosted by Nvidia. This achievement led to Hsu being named a “Jetson AI ambassador” for the country.

Competitive Landscape

MetAI operates in a competitive market, facing companies of varying sizes that offer digital twin technologies for manufacturing. Key competitors include Siemens Digital Industries, Dassault Systemes, Hexagon AB, Duality AI, and Intangles.

In the synthetic data space, numerous companies also exist, such as Sky Engine and Scale AI.

MetAI’s Unique Value Proposition

MetAI asserts that its approach is distinct from its competitors. “Unlike competitors that prioritize operational efficiencies or IoT integrations, MetAI leverages generative models and AI-driven layouts to create digital twins designed for physical AI training and implementation in real-world operations,” explains Yu.

This methodology not only accelerates digital twin creation but also ensures their direct applicability to advanced automation systems, like robotics, effectively bridging the gap between simulation and reality.

Synthetic Data Generation for Enhanced AI Training

MetAI further differentiates itself by generating artificial data within its AI-enabled digital twin environments. Yu emphasizes that this allows users to create synthetic data tailored to specific operational needs, thereby improving AI training and validation. “Instead of creating isolated datasets, MetAI builds dynamic virtual worlds (i.e., world simulators) — realistic virtual environments that operate exactly like the real world,” he stated.

Collaborations with Nvidia and Kenmec

The relatively young startup, established two years ago, offers a diverse product line encompassing vertical AI agents and digital twins. It currently serves a growing customer base and is achieving revenue generation through partnerships with companies operating within the manufacturing and automation sectors.

This year, the company anticipates revenue of $3 million from a single project, according to Yu. Income streams are comprised of project-based fees, product subscriptions, and licensing revenue from continuous development efforts.

“The integration of MetAI with NVIDIA Omniverse signifies a major advancement for industrial digital twins and the application of physical AI within simulations,” stated Nico Caprez, corporate development manager at Nvidia.

“Their capacity to develop scalable environments for AI training has the potential to establish a new benchmark across industries, including manufacturing and robotics.”

Digital Twin Development with Kenmec

In 2023, MetAI and Kenmec collaborated on the creation of digital twins specifically for automated warehouse environments.

MetAI’s technology reportedly reduced the simulation time required for warehouse digital twins from thousands of hours to a mere 3 minutes.

This substantial reduction in time translates to significant cost savings in both operational processes and verification procedures.

Future Expansion Plans

The recent funding will be utilized to bolster MetAI’s research and development team, accelerating product development and the implementation of its market entry strategies to address increasing demand.

Furthermore, the Taiwan-based startup intends to establish a U.S. presence and relocate its headquarters in the latter half of 2025, as confirmed by Yu to TechCrunch.

“Taiwan functions as our primary testing environment, allowing us to collaborate with leading Taiwanese companies to integrate specialized industry expertise into our models,” explained Liu.

“This ensures our solutions are both resilient and scalable.”

“Considering its scale and the demand for simulation-based solutions, we are targeting the U.S. market due to its high labor costs and complex operational landscapes.

Our expansion strategy centers on delivering both focused solutions and comprehensive end-to-end offerings, including SaaS platforms and vertical AI agents designed for swift deployment in practical, real-world applications within these industries.”