Depth Analysis of OpenMind: How People Collaborate with Bots?

Robot technology is rapidly advancing and attracting global follow. Walking with robotic dogs, or having humanoid robots assist with household chores, are no longer far away.

The next step is crucial. How do people collaborate with Bots? How do Bots coordinate with each other? This report will explore the answers to these questions through OpenMind.

Core Points

• OpenMind has developed the open-source runtime “OM1”. OM1 creates an environment where all Bots can communicate and collaborate freely, regardless of the manufacturer.
• OpenMind's blockchain network “FABRIC” has established a Bots identity verification, transaction record, and distributed verification system. FABRIC lays the foundation for an autonomous machine economy.
• OpenMind uses the ERC-7777 standard to define Bots behavior rules. OpenMind is collaborating with AIM Intelligence to develop a “Physical AI Security Layer”. These technologies will work together to prevent failures and block external attacks.

1. The growth rate of robot technology is beyond imagination.

Bots are no longer a distant future, nor do they only serve a few people.

Just a few years ago, Bots only appeared in laboratories or industrial settings. Now, they are stepping into our daily lives. People are walking robot dogs in parks or having humanoid Bots help with household chores; these are no longer scenes from science fiction movies.

Source: 1X Technologies

1X Technologies recently launched “Neo”, a home humanoid robot, making this reality more within reach. Consumers can now own a personal home assistant robot through a monthly subscription of $499 or a one-time payment of $20,000. The price is still high, but the message is clear: robotics has entered consumer homes.

Source: Made Visual Daily

Apart from Neo, global companies are accelerating innovation through fierce competition. Notable participants include Figure, Tesla, and Boston Dynamics from the United States, as well as Yushu Technology from China. Tesla plans to start mass production of its humanoid robot “Optimus” in 2026, with a price lower than its cars.

The robot industry is rapidly expanding into the consumer market. The once seemingly distant future has arrived faster than expected, opening the door to a new daily reality.

2. Robots in Daily Life: Possibilities and Limitations

What changes can robotic technology bring to our daily lives? Let's imagine a future where we live alongside Bots.

Neo is cleaning the house. Yushu's robot dog plays with the kids. Optimus goes to the supermarket to buy ingredients for dinner. Each robot works together, handling their respective tasks at the same time. Users experience a more efficient day.

Let's think further. What if Bots could collaborate to handle complex tasks?

Optimus is shopping at the supermarket. Neo checks the refrigerator and requests additional ingredients from Optimus. Figure adjusts the recipe based on the user's allergy information. Each Bots is connected in real-time, operating organically like a team. The user only needs to simply command: “I want to eat an omelette rice.”

But this is still a distant dream. Bots lack sufficient intelligence to flexibly respond to various situations. The bigger issue is that each bot operates within a closed system based on different technology stacks.

Robots from different manufacturers find it difficult to exchange data or collaborate smoothly. Photos can be transferred between iPhones via AirDrop, but AirDrop cannot be used with Samsung Galaxy phones. Robots face the same limitations.

Figure's Helix, Source: Figure

Of course, cooperation can be achieved under limited conditions, such as Figure's Helix: the same manufacturer, the same technology stack.

But reality is more complex. Take a look at the current Bots industry. A variety of Bots are flooding into the market like the Cambrian explosion.

In the future, users will choose various Bots based on their preferences and needs, rather than sticking to just one brand. Our households today prove this model. We choose Samsung refrigerators, LG washing machines, and Dyson vacuum cleaners.

Now imagine robots from multiple manufacturers working together in the same home. The kitchen robot cooks. The cleaning robot mops the floor. These two robots cannot share location information. Even if they share data, they cannot interpret it correctly. Their distance calculation methods and units of measurement are different.

They cannot track each other's movement paths. Collisions will occur. This is just a simple example. More Bots and complex tasks will amplify the chaos and collision risk.

3. OpenMind: Building a World of Bots Collaboration

Source: OpenMind

OpenMind was born to solve these problems.

OpenMind breaks the closed technology stack, pursuing an open ecosystem where all Bots can work together. This approach allows Bots from different manufacturers to communicate and collaborate freely.

OpenMind has proposed two core foundations to realize this vision. First, “OM1” serves as the open-source runtime for Bots. OM1 provides standardized communication methods, enabling all Bots to understand and collaborate with each other, despite different hardware.

Secondly, “FABRIC” operates as a blockchain-based network. FABRIC establishes a trusted collaborative environment among Bots. These two technologies create an ecosystem that allows all Bots, regardless of the manufacturer, to operate organically as a team.

3.1. OM1: Make Bots Smarter and More Flexible

As we have seen before, existing Bots are still trapped in closed systems, making it difficult for them to communicate with each other.

More specifically, Bots exchange information through binary data or structured code formats. These formats vary by manufacturer, hindering compatibility. For example, Company A's Bots represent location as (x, y, z) coordinates, while Company B defines it as (latitude, longitude, height). Even in the same space, they cannot understand each other's locations. Each manufacturer uses different data structures and formats.

Source: OpenMind

OpenMind solves this problem through the open-source runtime “OM1.” Think of it like Android; it runs on all devices regardless of the manufacturer. OM1 works in the same way, allowing all Bots to communicate in the same language regardless of hardware.

OM1 enables Bots to understand and process information based on natural language. OpenMind's paper “A Paragraph is Enough” explains this well. Communication between Bots does not require complex commands or formats. A context of natural language can achieve mutual understanding and collaboration.

Now let's take a detailed look at how OM1 works.

Source: OpenMind

First, the Bots collect environmental information from various sensor modules such as cameras and microphones. This data is input in binary format, but the multimodal recognition model converts it into natural language. VLM( visual language model) processes visual information. ASR( automatic speech recognition) processes audio. This generates sentences such as “A man is pointing at the chair in front” and “The user says 'go to the chair'”.

The converted sentences are aggregated through a natural language data bus. The data fusion engine weaves this information into a situational report and passes it to multiple LLMs. The LLMs analyze the situation through this report and decide the next actions of the Bots.

This approach has significant advantages. Robots from different manufacturers can collaborate seamlessly. OM1 forms a natural language-based abstraction layer on top of the hardware. Neo and Figure can understand the same natural language commands and perform the same tasks. Each manufacturer maintains its proprietary hardware and systems, while OM1 enables them to collaborate freely with other robots.

In addition to enabling cross-manufacturer collaboration, OM1 also integrates other open-source models as runtime modules, rather than competing with them. When precise operation is required, OM1 utilizes the Pi( physical intelligence) model. When multilingual speech recognition is needed, OM1 employs Meta's universal language ASR model. OM1 combines modules based on the situation, providing high scalability and flexibility.

The advantages of OM1 go beyond this. OM1 fundamentally leverages LLM. Bots are not just executing simple commands. They are capable of understanding contextual backgrounds and making autonomous decisions.

Pick up anything to demonstrate ( for ease of understanding using Figure's materials ), source: Figure

Let's look at a specific example. Multiple objects are placed in front of the Bots. Someone asks to “pick up an item related to the desert”. Traditional Bots would fail because “desert items” do not exist in predefined rules. OM1 is different. It understands conceptual relationships through LLM. It independently infers the connection between “desert” and “cactus”. It chose the cactus doll. OM1 lays the foundation for collaborative Bots and makes individual Bots smarter.

3.2. FABRIC: Connecting Distributed Bots into a Unified Network

OM1 makes Bots smarter and enables smooth communication between them. But aside from communication, there is another challenge. When different Bots collaborate, how do they trust each other? The system must verify who performed which tasks and whether they were completed correctly.

Human society regulates behavior through legal norms and ensures contract performance through agreements. These mechanisms enable people to conduct transactions and cooperate safely with strangers. The Bots ecosystem requires the same mechanisms.

Source: OpenMind

OpenMind addresses this issue through “FABRIC”, a blockchain-based network. FABRIC connects Bots and coordinates their collaboration.

Let's take a look at the core structure of FABRIC. FABRIC first assigns an “identity” to each Bot. Each Bot in the FABRIC network receives a unique identity based on the ERC-7777( human Bot social governance ).

Assigned identity Bots share their location, task status, and environmental information in real-time with the network. They simultaneously receive status updates from other Bots. Just like a situational board or mini-map in a management game, all Bots track each other's location and status in real-time through a shared map.

Simply sharing information is not enough. Bots may submit incorrect information. Sensor errors may occur and distort data. FABRIC utilizes the consensus mechanism of blockchain to ensure data reliability.

Consider a real scenario. Delivery Bot A collaborates with Warehouse Bot B to transport goods. Bot B reports that it is on the 2nd floor. Nearby sensor Bots and elevator Bots cross-verify Bot B's location. Multiple nodes validate the transaction on the blockchain. Multiple Bots operate in the same way. They confirm Bot B's actual location and reach a consensus. Suppose Bot B mistakenly reports the 2nd floor due to a sensor error, but is actually on the 3rd floor. The verification process will detect the discrepancy. The network records the corrected information. Bot A moves to the correct location on the 3rd floor.

The role of FABRIC is not limited to verification. FABRIC provides additional functionality for the upcoming machine economy. The first is privacy protection. Blockchain transparency ensures trust, but privacy is also crucial for operating a real robot ecosystem. FABRIC adopts a distributed structure, dividing subnetworks by tasks or locations, and connects them through a central server. This structure protects sensitive information. This solution is not perfect, but ongoing research will enhance privacy protection.

FABRIC also provides the machine settlement protocol (MSP). MSP automates hosting, verification, and settlement. When the system verifies that the task is completed, it automatically settles the payment in stablecoins and records all evidence on the blockchain. Bots will not just be cooperative entities that establish trust. They will become autonomous economic entities for trading.

4. If: Looking at the future daily life through OpenMind

4.1. A Brand New World: A Utopia with Bots

We have long dreamed of a “robot economy” where robots directly participate in economic activities. Robots make independent judgments, order goods, collaborate with other robots, and exchange value. OpenMind is now turning this dream into reality.

Source: OpenMind

What kind of daily life will unfold? Watch the demonstration video of OpenMind. You tell the Bots, “Please buy me lunch.” The Bots move to the store, confirm the order, pay directly with cryptocurrency, and bring the food back. On the surface, this seems very simple, but it is significant. Bots are no longer just executing commands in predefined environments. They are transforming into economic entities that judge and act independently.

Imagination can be further expanded. In addition to transactions between humans and Bots, there will also be transactions between Bots. For example, a household humanoid robot runs out of essentials while doing household chores. It independently orders products from a nearby supermarket Bot. An intelligent contract is automatically generated in the process. The supermarket Bot delivers the products. The household robot confirms the goods and settles the payment using stablecoins.

New forms of value exchange will emerge that have never existed before. Delivery Bots calculate the best route to the destination. They request real-time data from traffic Bots and pay a small fee in return. Even everyday small collaborations become transactions.

4.2. Dangerous World: Dystopia with Bots

Bots are no longer confined to science fiction movies. In China, consumers spend about $1,000 to purchase the robotic dog ( YuTree Go2), and about $12,000 to buy humanoid robots ( Engine AI PM01). Mass adoption is rapidly accelerating.

Source: WhistlinDiesel

The increase in the number of Bots in daily life is not the most important thing. The judgment ability of Bots is still limited. Safety has not been guaranteed. If a Bot misjudges a situation and makes a dangerous decision, it can cause direct harm to people. This harm could become a disaster, rather than just a simple accident.

OpenMind addresses this issue head-on. It assigns a unique identity to each Bot through the ERC-7777 standard and uses it as a safeguard. For example, a robotic dog receives the identity of “human friend and protector.” This identity prevents the Bot from attacking or harming people. The Bot always acts in a friendly and safe manner. The Bot continuously confirms its identity and role, and prevents inappropriate behavior.

OpenMind takes it a step further. They are collaborating with AIM Intelligence to develop a “physical AI safety layer.” This layer prevents Bots from hallucinating and defends against external intrusions and attacks. Consider an example. A Bot tries to move while holding a sharp object. A child is standing nearby. The system recognizes this as a “risk of injury” and immediately halts operations.

5. OpenMind: Building the Robot Society of Tomorrow

OpenMind has surpassed the research phase. It is ready to drive a substantial transformation in the Bots industry.

Founder Jan Liphardt, former professor of biophysics at Stanford University, is in a core position. He has studied the coordination and cooperation mechanisms between complex systems. He is now designing the structure for Bots to make autonomous judgments and collaborate. He leads the overall technology development.

Source: OpenMind

This technological leadership attracted $20 million in funding led by Pantera Capital. OpenMind has established a financial foundation for technological development and ecosystem expansion. It ensures the execution capability to realize the vision.

The market response is positive. Major hardware companies including Yushu, Deep Robotics(, Yujian), and UBTECH( are positioning OM1 as their core technology stack. The collaboration network is rapidly expanding.

However, challenges still exist. The FABRIC network is still in the preparation phase. Unlike the digital environment, the physical world presents more variables. Bots must operate in unpredictable real-world environments rather than controlled laboratories. Complexity increases significantly.

Nevertheless, Bots collaboration and security require long-term solutions. We need to follow how OpenMind addresses this challenge and what role it plays in the Bots ecosystem.