Noitom Robotics Named to The Information's 2026 List of the 50 Most Promising Startups Worldwide
Key Facts
· In The Information's "The 50 Most Promising Startups 2026" list released on November 11, Noitom Robotics ranks first in the Asia region.
· According to the list, Noitom Robotics currently has a valuation exceeding USD 500 million, cumulative equity financing of more than USD 50 million, and annual revenue surpassing USD 10 million.
· Noitom Robotics is dedicated to producing human data for robot learning. The company operates data factories in Shenzhen, China and Mountain View, California, and has partnered with nearly 100 companies in the Physical AI field.
Singapore, November 18, 2026 — Renowned global technology business media outlet The Information has officially released "The 50 Most Promising Startups 2026" list. Physical AI data company Noitom Robotics ranked first in the Top Asia Startups list.

Fig 1. The Information's 2026 List of the 50 Most Promising Startups Worldwide
List data shows that Noitom Robotics currently has a valuation exceeding USD 500 million, total equity financing of more than USD 50 million, and revenue of over USD 10 million.
Founded in 2013, The Information focuses on technology, artificial intelligence and finance, providing original exclusive reporting and in-depth analysis to business leaders and professional readers in more than 60 countries worldwide. The Information 50 list aims to identify early-stage startups with the greatest potential to become leaders in their respective segments, rigorously evaluating them on revenue scale, business model and growth prospects, with relevant financial data collected by journalists directly from founders and investors.
Dr. Dai Ruoli, Founder and CEO of Noitom Robotics, said:
"The Information ranks startups based on revenue and growth performance. Last year, robot R&D teams' demand for human motion data was still in the thousands-of-hours range; this year, that demand has surged to hundreds of thousands of hours. At the same time, another crucial question for these teams is whether embodied models can actually learn from this data. We are delighted that Noitom Robotics has been recognized by The Information. This inclusion also confirms the enormous commercial value of solving the two major pain points of 'data scale' and 'data learnability.'"
Not Building Robots, Focused on Producing the "Data" for Robot Learning
Noitom Robotics does not manufacture robots. The company focuses on producing high-quality training data that drives robot evolution — through self-built data factories and extensive offline collaboration scenarios, it systematically collects human motion, object interaction and daily behavior data, and provides standardized multimodal datasets to global robot本体 manufacturers and AI developers through its data platform ModalityNet (modalitynet.com).
After officially spinning off in 2025, Noitom Robotics leveraged its team's 13 years of motion capture technology expertise to rapidly build a global data production capacity and service network. The company now has over 100 employees, operates data factories in Shenzhen and Mountain View, California, and data collection centers in Taipei and Kuala Lumpur, and is actively expanding to other regions worldwide. To date, the company has established deep partnerships with nearly 100 companies and R&D institutions in the Physical AI field.
Continuing Milestones in Financing, Products and R&D Over the Past Year
The inclusion in The Information 50 stems from a series of consecutive milestones the company achieved in financing, products and R&D over the past year:
November 2025 — The company completed a Pre-A round of financing led by Alpha公社, with participation from Matrix Partners China (MPCi), Inno Angel Fund and other institutions.
December 2025 — The company completed a Pre-A+ round of financing led by Qiming Venture Partners, with participation from Alpha公社, Matrix Partners China, 5Y Capital and others.
June 2026 — Noitom Robotics officially launched its data platform ModalityNet, with three flagship dataset products simultaneously going live. It also announced the completion of a Pre-A++ round of financing totaling several hundred million RMB, with oversubscription from multiple leading industrial capital investors including Beijing AI Industry Investment Fund (managed by Shunxi), Shanghai AI Industry Series Fund, and Shenzhen Capital Group, as well as existing shareholders including Alpha公社, Legend Capital, 5Y Capital and Matrix Partners China.
August 2026 — The high-precision human interaction dataset HiPHI (High Precision Human Interaction) went live, open to the academic research community. HiPHI is the open research subset of the HiPHI-MOV (Motion with Object and Vision) corpus, available free for research use under the ModalityNet Open Research License (MORL v1.0), while the complete HiPHI-MOV corpus is commercially licensed through ModalityNet. Downloads on Hugging Face and the company's distribution channels have exceeded 15,000.
November 2026 — At the Conference on Robot Learning (CoRL 2026) held in Austin, three papers from the Noitom Robotics team were accepted to the main conference. At the same time, the company launched a new category of real-world human dataset, ITW-Ego Plus, and released the second open research dataset in the HiPHI series, HiPHI-Hand, which has been used to train NVIDIA's EgoHand model.

Fig 2. Data visualization of Noitom Robotics' real-world dataset ITW-EGO Plus
The complete The Information 50 list has been released to The Information subscribers. https://www.theinformation.com/projects/the-information-50-2026.
Noitom Robotics is a data company focused on Physical AI. The company is dedicated to building a "World Compiler" — transforming real-world human interaction into an infrastructure layer that machines can learn from, providing learnable representations for embodied intelligence. Its core implementation is ModalityNet (modalitynet.com), a large-scale, human-centered omnimodal data foundation built on three complementary data systems:
· HiPHI-MOV (High Precision Human Interaction — Motion with Object and Vision): collects high-fidelity full-body motion and mobile manipulation data;
· HiPHI-OM (High Precision Human Interaction — Omni-Manipulation): collects omnimodal interaction data including contact, force, tactile and object state signals;
· ITW-EgoVision & Ego Plus (In-The-Wild): collects natural human behavior data at scale in uncontrolled real-world environments.
Together, these three data systems address the core bottlenecks of Physical AI. Noitom Robotics' unchanging mission: making the physical world learnable.

- Alibaba Earnings: International Site GMV Continues Double-Digit Growth, Accio Paid Enterprises ...
- Alibaba Cloud External Revenue Grows 64%, Fastest Growth Rate in Nearly Six Years
- Focusing on AI Industry Foundation Building, Lingang AIGC Digital Content Creation Season Is Ongoing ...
- New Business Brought by AI | Rare Earth Nuggets X-Lab Annual Conference Recruitment Now Open ...