Nvidia is making significant strides into the realm of robotics and industrial AI applications with the introduction of the Omniverse platform and its subsequent Mega framework. As a component of the Omniverse Blueprint, Mega aims to develop digital twins for operating these applications. To further this initiative, Nvidia has also invested in startups specializing in digital twin technology.
Taiwan-based MetAI has developed a model that leverages AI and 3D technology to swiftly generate "SimReady" digital twins, capable of transforming CAD files into functional 3D environments within minutes.
Recently, Nvidia participated in MetAI's initial seed funding round of $4 million, marking the chip giant's first investment in a Taiwanese startup. Other investors in this round included Kenmec Mechanical Engineering, Solomo Technology, SparkLabs Taiwan, Addin Ventures, and Upstream Ventures.
The next generation of AI, known as generative physics AI, relies on physically accurate simulated environments to train and validate robots for autonomous systems and construct operational AI before deployment. MetAI believes that the digital twins it assists in creating will play a pivotal role in this process.
Focusing on advanced semiconductor factories, smart warehouses, and automation, MetAI specializes in AI-driven digital twins and generates synthetic data within AI-powered digital twin environments.
MetAI's co-founder and CTO Renton Hsu, with a background in 3D engineering and AI, first encountered digital twins when building AI software applications for enterprises, where they served as practical solutions for scenarios lacking sufficient data for system training. Hsu realized the same approach could be applied to 3D systems by combining 3D technology with AI, leading to the development of synthetic AI and 3D solutions. He then founded MetAI alongside co-founder Daniel Yu, experienced in digital transformation projects, and COO Dave Liu.
MetAI faces competition from both large and small companies developing digital twin technologies for manufacturing, such as Siemens Digital Industries, Dassault Systèmes, Hexagon, Duality AI, and Intagles. In the synthetic data space, competitors include Sky Engine and Scale AI.
MetAI distinguishes itself through its use of generative models and AI-driven layouts to create digital twins designed specifically for physical AI training and real-world implementation, accelerating digital twin creation while ensuring their direct applicability to advanced automation systems like robotics.
By generating artificial data in AI-powered digital twin environments, MetAI enables users to produce customized synthetic data tailored to specific operational requirements, facilitating AI training and validation. The company builds dynamic virtual worlds—world simulators—that mimic real-world operations.
This two-year-old startup offers a range of products from vertical AI agents to digital twins and collaborates with enterprises in manufacturing and automation industries, boasting some clients and revenue. According to Yu, MetAI expects to earn $3 million in revenue from a single project this year, derived from project income, product subscription fees, and ongoing development licensing fees.
Nvidia's Corporate Development Manager Nico Caprez stated that MetAI's integration with NVIDIA Omniverse represents a transformative advancement in industrial digital twins and physical AI simulations. MetAI's capability to create scalable environments for AI training may set new standards across various sectors, from manufacturing to robotics.
In 2023, MetAI collaborated with Kenmec to create digital twins for automated warehouses, claiming to reduce simulation time from thousands of hours to just three minutes, significantly cutting costs for operational and validation tasks.
With the latest funding, MetAI plans to expand its R&D team to accelerate development and execution of its go-to-market strategy to meet growing demand.
