Humanoid Bots + Crypto Assets: How does Reborn create the DePAI flywheel?

Author: brianbreslow, Hypersphere Ventures

Compiled by: Tim, PANews

Executive Summary

Humanoid general-purpose robots are rapidly moving from science fiction to reality. The continuous decline in hardware costs, the sustained growth in capital investment, along with technological breakthroughs in mobility and operational capabilities, are three key factors that are continuously merging and actively driving the next major platform iteration in the field of computing.

Despite the increasing commoditization of computing power and hardware, which brings low-cost advantages to robotic engineering, the industry is still constrained by the bottleneck of training data.

Reborn is one of the few projects that utilizes decentralized physical artificial intelligence (DePAI) to crowdsource high-precision motion and synthetic data, and to build foundational models for robotics, placing it in a uniquely advantageous position for advancing humanoid robot deployment. The project is led by a founding team with a strong technical background, with team members having academic research experience and professor backgrounds from the University of California, Berkeley, Cornell University, Harvard University, and Apple, demonstrating both exceptional academic standards and real-world engineering execution.

Humanoid Robots: From Science Fiction to Cutting-Edge Applications

The commercialization of robotics is not a new concept. The iRobot Roomba vacuum cleaner, which was launched in 2002 and is well-known to the public, and the recent Kasa pet camera and other household robots are all examples of single-function devices. With the development of artificial intelligence, robots are evolving from single-function machines to multifunctional forms, aiming to adapt to operations in open environments.

Humanoid robots will gradually upgrade from basic tasks such as cleaning and cooking to more complex jobs like reception services, firefighting, and even surgical operations over the next 5 to 15 years.

Recent developments are transforming humanoid robots from science fiction into reality.

Market dynamics: Over 100 companies are focusing on humanoid robots (such as Tesla, Utree Technology, Figure AI, Clone, Agile, etc.).

Hardware technology has successfully crossed the uncanny valley: the new generation of humanoid robots exhibits smooth and fluid movements, enabling them to achieve human-like interactions in real-world environments. Among them, the Yushu H1 can walk at a speed of 3.3 meters per second, far exceeding the average human walking speed of 1.4 meters per second.

(Note: The Uncanny Valley theory is a psychological theory that describes human emotional responses to non-human entities such as robots, dolls, virtual avatars, etc.)

The new paradigm of humanoid robot costs: expected to be below the salary level of American labor by 2032.

Development Bottleneck: Real-World Training Data

Despite the obvious favorable factors in the field of humanoid robots, the issues of low and insufficient data quality still hinder its large-scale deployment.

Other artificial intelligence entity technologies, such as autonomous driving technology, have essentially solved data issues through cameras and sensors mounted on existing vehicles. For example, autonomous driving systems like Tesla and Waymo can generate billions of miles of real road driving data. During this development stage, Waymo has real human monitors in the front passenger seat for real-time training when the vehicles are on the road.

However, consumers are unlikely to accept the existence of "robot nannies." Robots must have out-of-the-box high performance, which makes data collection before deployment crucial. All training must be completed before commercial production, and the scale and quality of the data remain ongoing challenges.

Although each training mode has its own scale units (for example, large language models use tokens, image generators use video-text pairs, and robotics use motion segments), the comparison below clearly reveals the magnitude gap in data availability faced by robotics:

The training data scale of GPT-4 exceeds 150 trillion text tokens.

Midjourney and Sora utilize billions of labeled video-text pairs.

In contrast, the largest robot dataset contains only about 2.4 million interaction records.

This gap explains why robotics technology has not yet achieved a true foundational model like large language models; the key lies in the incomplete data foundation.

Traditional data collection methods struggle to meet the scalability requirements for training data of humanoid robots. Existing methods include:

Simulation: Low cost but lacks real boundary scenarios (the gap between simulation and reality)

Internet video: Unable to provide the proprioceptive and force feedback environment necessary for robot learning.

Real-world data: Although accurate, it requires remote control and human closed-loop operations, which leads to high costs (over $40,000 per robot) and lacks scalability.

Training models in virtual environments is cost-effective and highly scalable, but these models often struggle to be deployed in the real world. This issue is referred to as the Sim2Real gap.

For example, a robot trained in a simulated environment may easily pick up objects that are perfectly lit and have smooth surfaces, but when faced with a chaotic environment, uneven textures, or various unexpected situations that humans take for granted in the real world, it often finds itself at a loss.

Reborn provides an economical and efficient method for quickly crowdsourcing real-world data, aiding in the enhancement of robot training and addressing the "Sim2Real" challenge.

Reborn: The Full-Stack Vision of Decentralized Entity AI

Reborn is building a vertically integrated software and data platform for embodied intelligent robot applications. The company's core goal is to address the data bottleneck issue in the field of humanoid robotics, but its vision goes far beyond that. By combining self-developed hardware, multi-modal simulation infrastructure, and foundational models, Reborn will become a full-stack driver for achieving embodied intelligence.

The Reborn platform is built on a rapidly expanding ecosystem of augmented and virtual reality games, starting with a proprietary consumer motion capture device called "ReboCap". Users can provide high-fidelity sports data in exchange for online incentives and rewards to promote the sustainable development of the platform. To date, Reborn has sold more than 5,000 ReboCap devices, has 160,000 monthly active users, and has established a clear growth path to surpass 2 million users by the end of the year.

Reborn's support for data collection significantly outperforms other solutions.

Notably, this growth is entirely driven by natural development: users are attracted by the entertainment value of the game, while streamers utilize ReboCap to achieve real-time motion capture of their digital avatars. This spontaneously formed virtuous cycle enables scalable, low-cost, high-fidelity data production, making the Reborn dataset a training resource eagerly adopted by top robotics companies.

The second layer of the ReBorn software stack is Roboverse: a multimodal data platform that unifies fragmented simulation environments. Currently, the simulation field is highly fragmented; tools such as Mujoco and NVIDIA Isaac Lab operate independently, each with its own advantages but lacking interoperability. This fragmented situation slows down the development process and exacerbates the gap between simulation and reality. Roboverse creates a shared virtual infrastructure for developing and evaluating robotic models by standardizing multiple simulators. This integration supports consistent benchmark testing, significantly enhancing the system's scalability and generalization capabilities.

Roboverse enables seamless collaboration. The former collects real-world data on a large scale, while the latter builds simulation environments to drive model training. Together, they showcase the true power of the Reborn distributed physical intelligence network. The platform is creating a developer ecosystem for physical artificial intelligence that goes beyond mere data acquisition, and its functions have extended into actual model deployment and commercial licensing.

Reborn Basic Model

The most critical component in the Reborn tech stack may be the Reborn foundational model (RFM). As one of the first foundational models for robots, this model is being developed as the core system of emerging physical artificial intelligence infrastructure. Its positioning is similar to traditional large language foundational models, akin to OpenAI's GPT-4 or Meta's Llama, but aimed at the robotics field.

Reborn Technology Stack

The three core components of the Reborn tech stack (ReboCap data platform, Roboverse simulation system, and RFM model authorization mechanism) together build a solid vertical integration moat. By combining crowdsourced motion data with a powerful simulation system and model authorization system, Reborn is able to train foundational models with cross-scenario generalization capabilities. This model can support diverse robotic applications in the industrial, consumer, and research fields, achieving generalized deployment under massive and varied data.

Reborn is actively promoting the commercialization of its technology, launching paid pilot projects with Galbot and Noematrix, and establishing strategic partnerships with Unitree, Booster Robotics, Swiss Mile, and Agile Robots. The humanoid robot market in China is experiencing rapid growth, accounting for approximately 32.7% of the global market. Notably, Unitree Technology holds over 60% of the global quadruped robot market and is one of six Chinese manufacturers planning to produce more than 1,000 units (humanoid robots) by 2025.

The role of cryptocurrency technology in the physical artificial intelligence technology stack.

Cryptography is building a complete vertical stack for artificial intelligence in the physical world.

Reborn is a leading embodied artificial intelligence cryptocurrency project.

Although these projects belong to different layers of the physical artificial intelligence stack, they share one commonality: they are all 100% DePAI projects. DePAI creates an open, composable, and permissionless scaling mechanism throughout the technology stack through token incentives, and it is this innovation that makes the decentralized development of physical artificial intelligence a reality.

Reborn has yet to issue tokens, and the organic growth of its business is even more valuable. When the token incentive mechanism is officially launched, network participation will accelerate as a key part of the DePAI flywheel effect: users can get incentives from the project team when they purchase Reborn hardware devices (ReboCap collectors), and robot R&D companies will pay contribution rewards to ReboCap holders, and this dual incentive will drive more people to purchase and use ReboCap devices. At the same time, the project team will dynamically incentivize high-value customized behavioral data collection, so as to more effectively bridge the technology gap between simulation and real-world applications (Sim2Real).

The DePAI flywheel of Reborn is running.

The "ChatGPT moment" in the field of robotics will not be triggered by the robotics companies themselves, as hardware deployment is far more complex than software. The explosive growth of robotics technology is inherently limited by cost, hardware availability, and deployment complexity, obstacles that do not exist in purely digital software like ChatGPT.

The turning point for humanoid robots was not how impressive the prototype was, but how affordable the cost was, as it did when smartphones or computers were commonplace. When costs come down, hardware becomes the ticket in, and the real competitive advantage lies in the data and models: specifically, the scale, quality, and diversity of the motion intelligence used to train machines.

Conclusion

The robot platform revolution is unstoppable, but like all platforms, its large-scale development relies on data support. Reborn, as a high-leverage bet, firmly believes that cryptographic technology can fill the most critical gap in the AI robot technology stack: its robot data solution DePAI is cost-effective, highly scalable, and modular. As robot technology becomes the next frontier of AI, Reborn is turning the general public into "miners" of action data. Just as large language models need text annotations for support, humanoid robots require massive sequences of actions for training. Through Reborn, we will break through the last bottleneck and achieve the leap from science fiction to reality for humanoid robots.