WABO Technology has completed a new funding round backed by Vertex Ventures, Hongtai Capital, Chengdu Future Industry Fund under Chengdu Science and Technology Innovation Investment Group, and Huifengda Capital. In the original report, Vertex Ventures was highlighted for another well-known bet in its portfolio: an early investment in Unitree Robotics.
The financing led to an interview between Chinese venture outlet PEdaily and WABO founder Zhang Xingzhi. Zhang described a path that started years earlier, when a scene in Avatar left a strong impression on him: a human controlling another body through a device worn on the head. That image, he said, set off a pursuit that lasted more than a decade and eventually pushed him to launch WABO last year, after deciding the timing for the industry had become viable.
From human-computer interaction research to a brain-computer interface startup
Zhang’s core question has been consistent for years: how can human goals, choices, and movement intention be mapped onto another body so that human capability can extend into the physical world. He began working on human-computer interaction as an undergraduate.
At that stage, brain-computer interface technology was still far from commercial reality. Zhang first moved through industry, working close to manufacturing, consumer electronics, and supply chains. He later brought those industrial questions back into academia and continued his graduate studies at Tsinghua University, where he pursued both a master’s degree and a doctorate. The direction stayed the same through both phases: finding a lower-burden, more natural way for machines to understand and execute human intent.
He sees 2025 as the point when several pieces finally lined up. China’s 15th Five-Year Plan placed brain-computer interfaces among future industries for forward-looking development. At the same time, flexible electrodes, low-power chips, on-device AI, and spatial computing terminals improved enough to make non-invasive brain-computer interfaces look commercially feasible.
Against that backdrop, Zhang founded WABO at the end of last year. The company is based in Chengdu and follows a non-invasive route. Because it was set up around the same time as Gestalt Technology, another Chengdu startup pursuing an ultrasound brain-computer interface route, the two were quickly grouped together by outside observers as Chengdu’s “brain-computer twins.”
WABO is targeting the model layer, not only brain-computer interface hardware
The report says WABO is taking a different approach from many companies focused on a single medical use case or a specific piece of brain-computer interface hardware. Zhang describes the company more as a model company.
Artificial intelligence can already process language, images, audio, and motion after a decade of progress. What machines still lack, in Zhang’s view, is one crucial category of data: what a person intends to do before the action actually happens.
A camera can see someone pick up a cup, but it cannot directly tell whether that person wants to drink water or put the cup back. A robot can imitate a motion path, but it still struggles to understand when a human has confirmed a decision, when hesitation appears, when a correction is made, or when the machine should stop immediately.
That is the problem WABO says it wants to solve. The team is building what it calls a “human neural intention model,” a neural signal representation and decoding system built for real tasks. In a defined task, environment, and time context, it combines EEG, EMG, behavior, vision, device status, and execution results to judge the user’s current goal, action tendency, confirmation, correction, stop, or takeover intent, then turns that into a standardized intention vector that machines can read.
Zhang said the model has three key features. First, it is task-conditioned, meaning the model knows what the user is doing and what the legal options are. Second, it is designed to generalize across users, time periods, and devices, reducing repeated training and long calibration cycles. Third, it is closed-loop. In his words, a reliable brain-computer interaction system must not only recognize explicit commands correctly, but also stay silent when the user is distracted, blinking, speaking, moving, or simply not trying to control anything.
For Zhang, the shift from lab demo to broad deployment depends on that distinction. Machines need to learn not just when to act, but also when not to act.
He links that capability to a much wider application range than today’s early-stage market suggests, from medical rehabilitation and smart home systems to embodied intelligence. The report also notes that the more speculative vision represented by Avatar remains distant, but Zhang sees continuous model iteration as the path that may gradually move those ideas closer to reality.
Seed financing arrived in less than five months, then the angel round began
Zhang said he left little room for retreat when he decided to start the company. Once he concluded that the industry window was opening, he put in his personal savings and went all in.
The team formed quickly. WABO began with Zhang alone, then soon attracted like-minded recruits. The core team’s average age is under 30, according to the report, and is described as combining a strong engineering temperament with full-stack capability.
Senior research support was added as well. On the underlying R&D side, WABO has brought together brain-computer interface, neural engineering, and neural intention research talent from institutions including Tianjin University and Fudan University. The company says this interdisciplinary scientist team is working on the foundational research, data system, and closed-loop validation for the human neural intention model.
Investors followed fast. In less than five months of actual operations, WABO completed a seed round from Ceyuan Capital. Almost at the same time the seed round closed, the current angel round was launched. The report says at least one investment firm, trying to catch the window, moved both its post-investment and risk-control teams on site and completed due diligence and internal decision-making within a week.
Vertex Ventures, one of the main investors in the latest round, was drawn to WABO’s system-level capability across data collection, signal processing, device control, model training, and engineering delivery. The report says that profile stood out to a firm that has been actively deploying capital in Physical AI in recent years.
The broader investor takeaway, according to the article, is that scarcity in this market is not defined by a single hardware metric. The scarce asset is a complete system that can bring neural signals into real-world scenarios in a stable way and keep converting operating results back into model capability.
Brain-computer interfaces as an entry point into robots, drones, and space systems
The report frames brain-computer interfaces as only one side of the stack. On the other side could sit drones, robotic arms, humanoid robots, and other physical devices. That helps explain why investors who previously focused on embodied intelligence and commercial space have also backed WABO. Their bet, as presented in the article, is not tied to one brain-computer hardware product but to a reusable layer of human intention representation that can work across devices.
Space is one of the examples cited. In microgravity, traditional interaction methods based on hand gestures, voice, or physical interfaces face clear constraints. Brain-computer interfaces could become a more natural and more important way to interact. For astronauts on long-duration space station missions, the report says microgravity, confinement, and radiation can create neural fatigue and cognitive decline. A space-oriented brain-computer interface could monitor brain states in real time and dynamically regulate cognitive load, making it part of a next-generation human-machine integration system for crewed spaceflight.
At the global level, China and the US are both described as working on orbital brain-computer experiments. Even so, commercially deployable solutions that also meet space requirements such as miniaturization, low power consumption, radiation resistance, and high stability remain rare.
The report adds that WABO is already working with Tianjin University in this advanced direction. Tianjin University’s brain-computer interface team is described as an early mover in China’s space brain-computer interaction field. Its self-developed equipment previously flew on the Tiangong-2 and Shenzhou-11 missions, where it completed what the report calls the first in-orbit space brain-computer interaction scientific experiment in human history.
That frontier work in “space brain-computer” also caught the attention of Hongtai Capital, which has been active in commercial space investments in recent years.
Although WABO is a newer entrant, the company is described as evolving quickly. The original article says its control targets, smoothness, and accuracy are improving on a weekly cycle, and that investors see a new version of the company each time they visit. It also says several investors have already approached WABO in hopes of securing allocation in the next round.
Dual-headquarters plan takes shape in Chengdu and Shanghai
Another milestone is now approaching. WABO has completed landing coordination with the Minhang district government in Shanghai and with “Brain-Intelligence World,” and plans to formally enter through a brain-computer interface innovation center project incubated by Fudan University.
The report describes Brain-Intelligence World as a key Shanghai initiative and the country’s first future-industry cluster dedicated to brain-computer interfaces, already bringing together leading domestic companies in the field. After the move, WABO plans to adopt a dual-headquarters setup spanning Chengdu and Shanghai.
The two locations will split roles. Shanghai will focus on advanced research, neural intention models, closed-loop neural modulation, and coordination across industry, academia, research, and medicine. Chengdu will focus on hardware and product development, supply chains, engineering delivery, and scaled deployment in application scenarios. The company’s plan is to build frontier research capability and industrial engineering capability in parallel.
Industry financing is rising, but long-term competition may come down to data, models, and deployment loops
WABO’s financing and expansion are presented in the article as a small but revealing example of a broader shift in brain-computer interfaces. Based on incomplete statistics cited by the report, China recorded more than 30 financing events in the brain-computer interface field in the first five months of 2026, with total funding above RMB 4.5 billion. That figure exceeded the scale seen in all of 2025.
One investor following the sector said that “almost every direction can raise money.” Zhang said the change is obvious from the front line of the industry. In the past, investors mostly focused on the medical-device identity of brain-computer interfaces or on isolated research achievements. Now investors from embodied intelligence, AI, consumer electronics, and commercial space are starting to reassess the cross-sector value of the technology.
His view is that future competition will not center only on one hardware device. The decisive capability will be the ability to keep generating data, train models, and move into more physical scenarios.
WABO’s answer is to use neural signals such as EEG and EMG as the data entry point, align those signals with user behavior, scene information, and execution results in defined task environments, and build a continuously trainable neural intention model. Through what it calls WABO Interface, the company then converts human intention into control signals that can be used by physical endpoints such as robots, XR systems, smart home devices, and spacecraft.
Within that system, products such as Lingxi and Dreamweaver are not the endpoint, according to the report. They are data entry tools for real-world use. The terminal reaches the user, the user generates feedback in real scenarios, data pushes the model to iterate, and stronger model capability then improves the product experience, forming a long-term loop of “neural data entry - intention model - physical control - data return.”
That loop is also what investors appear to value most. The report says WABO is less interested in collecting isolated EEG curves than in understanding the task relationship behind the data: in what scenario the user generated a given neural signal, how the model interpreted the intent, whether the machine succeeded after execution, and whether the user confirmed, corrected, or took over. In that framework, the valuable unit is not a standalone waveform but a complete chain linking task, intention, action, and result.
Asked about future industry structure, Zhang said he does not expect a winner-takes-all outcome built around one technical route. Invasive and non-invasive systems, consumer neural interaction, core components, and data-model players may all find positions in the market. Before that happens, though, the field still has to clear obstacles around scenario generalization, cross-user and cross-subject transfer, safety boundaries, data standards, cost, mass production, and commercial demand.
Zhang’s view is that the next three years will bring real differentiation to the brain-computer interface industry. The companies that remain, he said, will be the few that can genuinely close the loop across data, models, and scenarios.