Forwarding the original title: “Investment Thesis: Amiko - Your Personal AI, Controlled and Evolved by You”

We have noticed Humane Pin and recognize its vision: to create a wearable device that seamlessly integrates into daily life as a super AI assistant. However, we also observe a core gap: the market does not lack functional AI assistants, but rather lacks a true personal digital companion, akin to the relationship depicted in the movie “Her.” We are more inclined towards products that resemble “electronic pets (Tamagotchi)” with emotional connections, a digital existence that evolves alongside the user as they grow. This deep personalization endows the product with significant companionship value, rather than just technical execution capability. Furthermore, we firmly believe that users’ data ownership should be firmly held in the hands of the users themselves, rather than being controlled by third-party platforms.

AmikoAs a leading project that aligns with this trend, it occupies an excellent position in the next stage of personal AI evolution. Against the backdrop of the rapid expansion of the global dedicated AI hardware market—where the global wearable AI device market is projected to grow from $21.2 billion in 2023 to $221 billion in 2032, with a compound annual growth rate of nearly 30%—Amiko stands out with its differentiated positioning. Unlike large tech companies (such as the collaboration between OpenAI and Apple, and the new device designs led by Jony Ive) that emphasize centralized control and cloud dependency, Amiko adopts a decentralized architecture that gives users complete ownership and control over their data. This core principle underpins its full-stack technology solution: emphasizing user sovereignty, a local-first computing model, and an AI relational system with emotional understanding capabilities. In an era where privacy anxiety is on the rise, Amiko builds a strong differentiated competitive barrier through its high focus on user security and emotional connection.

Furthermore, Amiko is not a single AI agent but serves as a coordination hub between hardware and software, accommodating multiple AI agents to operate collaboratively, truly integrating into the user’s daily life. For example, users can primarily interact with Eliza while also being able to integrate other personalized agents, collectively forming a flexible and diverse AI ecosystem. This system effectively addresses the pain points of high latency, poor personalization, and weak privacy associated with centralized cloud AI, providing a faster, safer, and more responsive AI experience.

Problems and Solutions

Current mainstream AI devices (especially those launched by large enterprises) have several structural flaws: they are highly dependent on cloud infrastructure, produce uniform outputs, lack continuous memory functions, and miss personalized expression and companionship, remaining in the role of “tools” and struggling to establish user stickiness. The more fundamental issue lies in the lack of emotional interaction: what users truly desire is intelligent hardware that can create emotional connections. This is precisely why the “electronic pet-style evolution” model meets user needs—users hope to cultivate an AI companion that understands them and evolves gradually.

Humane AI Pin provides a representative counterexample. Despite attracting industry attention and capital support, its total sales over its lifecycle are just above $9 million, with over $1 million worth of products returned. In critical months, the return volume even exceeded sales volume, and the actual market inventory is only about 7,000 units, far below the original target of 100,000 units. This indicates that if a closed, centralized AI device lacking emotional experience attempts to replace smartphones, it is difficult for the product to form a deep connection with users.

Even more advanced strategies, such as OpenAI integrating with Apple’s systems (like Siri/iOS) or collaborating with Jony Ive to develop new AI hardware, have validated the market’s potential, but still face critical shortcomings: these solutions remain cloud-centric, user data is easily controlled by the platform, and AI agents lack personalized evolution and ownership. Such strategies emphasize computing power and functional output, but fail to provide the user sovereignty and emotional evolution path that Amiko highlights.

In contrast, Amiko offers a complete and systematic alternative: a personal AI framework built on decentralized computing and hardware autonomy. Its initial form, Kick, will ensure data privacy and processing efficiency through a distributed computing network, and subsequently, Brain will serve as the household hub to achieve complete local AI inference processing, truly realizing device autonomy and data sovereignty. This path not only takes into account market rhythms but also allows for rapid validation of product experience, establishing a paradigm for the widespread deployment of local AI devices in the future.

Product Composition

• Hardware: including the portable device Kick and the home device Brain, serving as the physical carriers and computing units for AI agents.
• Platform software: including local inference engine, structured memory system, and agent coordination framework, to build an AI experience with emotional perception capabilities.
• Companionship: Supports various AI agents with emotional recognition and expression capabilities, such as Eliza, to achieve personality connection and companionship interaction.
• Autonomy: Supports a digital twin that grows together with users, continuously evolving without relying on specific agents, forming a unique digital individual.

Technology and Architecture

The Amiko architecture is centered around user sovereignty and data privacy, resulting in significant advantages over centralized platforms. In the Kick phase, AI reasoning is conducted based on a decentralized network, ensuring that data processing does not leave the local environment; in the Brain phase, all data will be processed on the device side, truly achieving “local autonomy.”

Core technology components include:

• Local language model: Kick runs a lightweight model, while Brain runs a fine-tuned model, enabling edge-side deployment.
• Memory Engine: Equipped with structured long-term memory capabilities, simulating human cognitive processes, private local storage.
• Multi-Agent Coordination Layer: Supports multiple AIs working collaboratively, with the main agent scheduling defined by the user.
• Personalized Agreement: Supports users in defining the tone style, interaction method, and level of autonomy for the agent.
• MCP Protocol: Ensures that agents have access rights to local applications and tools while ensuring that data is not leaked.

Core Ternary Structure

Amiko is built on three core elements, which interact and collaborate to achieve a highly personalized AI user experience.

• AI Agents: Diverse AI partners like Eliza, AIxBT, etc., with different personalities and functionalities, provide users with task management, communication support, and goal assistance.
• Digital Twin: A personalized AI avatar generated during continuous interaction, deeply understanding user preferences, habits, and behavioral patterns, truly representing the user, and owned by the user with all its training data, capable of collaborative use across agent platforms.
• User Entity: Users have full control and can define the boundaries of their twin entities, select the composition of their agents, ensuring that the system always serves individual values and needs.

Market Opportunities

With the migration of AI products to the hardware side, users’ demand for personalization, privacy control, and long-term companionship is growing. The multi-agent, decentralized, privacy-first architecture built by Amiko has the potential for cross-scenario expansion, covering core needs such as production efficiency, health management, learning, and social communication. Compared to other AI systems that rely on existing platforms or centralized solutions, Amiko offers greater flexibility, adaptability, and the ability to build user loyalty.

Development Path and Future Potential Amiko’s development path is clear. Initially, it verifies interaction experience and user feedback through Kick, guiding the preliminary construction of the ecosystem. Subsequently, it achieves full-end autonomy and high-performance local model deployment through Brain, fully supporting the operation of digital twins. At the same time, it will launch supporting application function iterations, open up the agency market mechanism, and introduce ecological incentives and multi-level data protection protocols. This development rhythm is highly aligned with the evolution trend of AI hardware, providing a practical foothold for future personal AI operating systems.

Conclusion

At Ryze Labs, we focus on investing in underlying technological architectures that can reconstruct the relationship between humans and machines as well as computing boundaries. Amiko’s product vision, technical path, and emotional philosophy together form the prototype of the next generation personal AI system. Its team has demonstrated efficient execution, a clear roadmap, and adherence to principled values (privacy protection, user control, long-term companionship). In the current market trend of significantly rising demand for data sovereignty and AI personification, Amiko provides an alternative path distinct from existing tool-based AIs or corporate assistants. With edge computing architecture, emotional AI framework, and decentralized control concepts, Amiko has built a dual moat of emotional barriers and technological barriers. We believe Amiko is redefining the personal AI category: emphasizing presence over prompts, context over convenience, and relationship building over functionality. In a world filled with corporate assistants, Amiko is the true AI companion that belongs to the user—growing together with them.

Statement:

1. This article is reproduced from [TechFlow] , Original title “Investment Thesis: Amiko - Your Private AI, Controlled by You and Evolving Together”, copyright belongs to the original author [ TechFlow], if there are any objections to the reprint, please contactGate Learn TeamThe team will process it as soon as possible according to the relevant procedures.
2. Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Other language versions of the article are translated by the Gate Learn team, unless otherwise mentioned.GateUnder these circumstances, it is prohibited to copy, disseminate, or plagiarize translated articles.