On a future day when AI media has become a natural part of everyday life, I send this record to you — the one who continues to devote day and night to unraveling the origins of matter through materials science research.
Through a new concept called Ontology Media-Lab, we explore the intersection where media, materials science, and AI converge. Through technological advancement and the power of media, we seek to interpret our world through a new lens, connect the diverse contents within it, and guide people’s curiosity toward the knowledge we pursue — the deeper meaning behind information — through ontology-based structures. This is the first step toward a complex research endeavor that weaves all these elements together and transforms them into value for the benefit of everyone.
AI possesses the remarkable ability to connect fragmented information into an organic semantic network. In this process, materials science experts and AI do not exist in opposition to one another; instead, they collaborate to create knowledge and write together. This is how a new form of media can be realized. Through the establishment of ethical principles that define the coexistence of humans and AI, this article aims to present a vision of a future intelligent media ecosystem — one built upon the philosophy of meaningful connections, shared value, and benefits that extend to all.
Topic Clusters: Beginning with This Fundamental Concept
A topic cluster is an information-structuring strategy that organizes related content around a central theme. A core page, known as pillar content, provides a comprehensive overview of the main topic, while multiple interconnected cluster contents explore specific subtopics in greater depth. This approach not only enables search engines to better understand the expertise and authority of the content but also allows users to navigate complex information more systematically.
MatOnDX.org has initiated the development of a blog platform based on this topic cluster structure, focusing on emerging technologies and key issues in the field of data-driven materials science. In the future, this structure will evolve beyond simple textual connections toward a contextual network that logically organizes relationships, hierarchies, and dependencies among different pieces of knowledge, thereby deepening the meaning and value of information.
As the content within the media platform continues to expand, this network will evolve into a more sophisticated knowledge structure — similar to an encyclopedia or wiki — where vast amounts of information are precisely interconnected. Through this intelligent media ecosystem, users will be able to navigate beyond fragmented information and efficiently reach the essential knowledge they seek.
However, these interconnected structures and AI technologies alone cannot create a trustworthy knowledge ecosystem. Even the most sophisticated topic clusters and semantic networks may become channels for spreading errors if the accuracy, depth, and reliability of the underlying content are not ensured. In an era where generative AI is becoming an important tool for media creation, a critical challenge is determining how AI-generated outputs can be validated and integrated with expert knowledge to produce reliable and meaningful content.
MatOnDX.org views AI not merely as a writing tool, but as a collaborative partner in knowledge creation alongside domain experts. At the same time, we recognize the importance of establishing responsibilities, ethical principles, and quality standards that experts must uphold when working with AI. Based on this philosophy, we propose the following fundamental principles that AI-driven media should follow as humans and AI create knowledge together.
Guidelines for Collaborative Writing between Materials Experts and AI: Media Standards
To realize our vision, clear standards are essential for the collaboration between AI and materials science experts. While AI is a powerful partner in knowledge production, the ultimate responsibility for accuracy and reliability lies with the domain expert. Therefore, all content must be finalized based on expert verification and judgment. Collaboration between AI and humans must transcend mere productivity gains to become a process of accumulating trustworthy knowledge. The following principles represent the core philosophy of MatOnDX.org’s AI-driven media and the standards for knowledge production that we will continue to evolve.
1. Principle of Trustworthiness
AI is merely a collaborator in knowledge production; the domain expert remains the primary subject of responsibility. Accordingly, all AI-generated content must be thoroughly reviewed and fact-checked by the author, guided by their conscience and professional expertise, regardless of external demands or verification requirements. MatOnDX.org does not consume AI-generated results as-is but strives to be a reliable media platform through responsible expert verification.
2. Principle of Knowledge Value
High-quality AI-assisted content is the result of meticulous prompt design, iterative verification, and professional judgment. MatOnDX.org is committed to systematically organizing and developing the prompts, reasoning methods, and domain knowledge accumulated through this process. Our mission is to move beyond simple information delivery and provide knowledge that readers can trust. In doing so, we aim to reduce the time and cost associated with AI utilization and create an environment where expert knowledge is more efficiently accessible to everyone.
3. Principle of Originality
Collaboration with AI involves more than just generating drafts or refining sentences. To enhance the integrity of the work, we repeatedly design prompts for specific sub-topics, integrate partial reasoning results, and continuously reflect expert interpretation and judgment. MatOnDX.org does not pursue volatile content created through one-off conversations or by delegating the entire writing process to AI. Every piece of content must embody the domain expert’s original perspective and deep context, serving as a product of high-level reasoning and knowledge generation based on semantic connections.
4. Principle of Continuous Evolution
AI technology and the social environment are in a state of constant flux. MatOnDX.org will continuously improve and publicly disclose these AI media writing guidelines in response to these changes, fostering a culture of responsible knowledge production where experts and AI grow together.
5. Principle of Semantic Connectivity
No content should remain as an isolated piece of information; instead, it must be capable of connecting with other knowledge through ontologies and semantic networks. MatOnDX.org views media not as a final destination for knowledge, but as a starting point for new connections. By simultaneously building content that humans understand and semantic structures that AI can interpret, we strive to create a knowledge ecosystem that benefits everyone.
Content accumulated based on these principles is not merely consumed as individual posts; rather, each piece of writing is interconnected and refined in meaning, eventually evolving into a comprehensive system of knowledge. MatOnDX.org focuses on a fundamental question: can media go beyond simply utilizing ontologies to become a space for knowledge production that generates its own? This represents the possibility of media evolving into a platform where AI and humans co-accumulate meaning and form entirely new knowledge.
Does the Ontology Discovered within Media Benefit People?
MatOnDX.org perceives the vast amount of content accumulated over time on specific topics as a single, unified context. By systematically analyzing the concepts and relationships within this content and refining recurring semantic structures, we believe it is possible to discover a unique knowledge structure—an Ontology—native to that media. This is not just a collection of concepts or a data model; it is the “thought structure” of the media, reflecting a specific way of perceiving and understanding a field. We define this as the Media Lens.
The Media Lens forms a unique perspective through which to view the world. Through this lens, individual pieces of information and content become semantically linked, and fragmented data transforms into new context and knowledge within a semantic network. Ultimately, media evolves from a space of information consumption into a platform where humans and AI interpret the world together, connecting meanings and generating new insights. We believe this Media Lens will serve as the core foundation for making specialized knowledge more accessible and creating a knowledge ecosystem that benefits everyone.
Can We Discover an AI Persona in the Future of Materials Science?
One of the most intriguing questions MatOnDX.org seeks to explore is whether AI, when writing based on an ontology reduced from media content, can reproduce the original author’s unique perspective and way of thinking. This goes beyond merely imitating a writing style or mimicking expressions. Within a vast body of work accumulated over time lies not only an expert’s specialized knowledge but also their unique way of viewing concepts, interpreting problems, and connecting disparate pieces of information. By systematically extracting and refining these semantic structures through ontologies and semantic networks, we are focused on the possibility of implementing an AI Persona that reflects an individual’s expertise and philosophy.
Such an AI Persona is not simply a tool for automatically generating new posts. Instead, it can become a new subject of knowledge production that continuously inherits and expands the unique insights and expertise a domain specialist has accumulated over years of research. If an expert’s experience, judgment, and academic philosophy can be structured through semantic connections—allowing AI to interpret new problems and regenerate knowledge based on that foundation—human intelligence will no longer be confined to individual documents but will be continuously expanded and restored. We view this possibility as a vital research theme for future intelligent media.
Ultimately, our goal is to build a knowledge ecosystem that benefits everyone, grounded in the Media Lens and the AI Persona. By interpreting the world through diverse perspectives via the Media Lens and connecting accumulated content through ontologies and semantic networks, we aim to develop high-level expertise into a knowledge asset that anyone can understand and utilize. This connectivity transcends simple information delivery, creating a virtuous cycle where new knowledge is generated and both experts and AI grow together.
Furthermore, the knowledge network built on these ontologies serves as the foundation for connecting the value created by data, AI technology, educational content, and media activities within a single ecosystem. In this ecosystem, we strive for a structure where knowledge producers and contributors are justly recognized, and the value and revenue generated through media are transparently shared with all members. We view media not as the final destination of knowledge, but as the starting point for new connections and creation. A future where human-readable content and AI-understandable semantic structures grow together, returning that accumulated intelligence to society as a whole—this is the future knowledge ecosystem that MatOnDX ultimately envisions.
Introduction to OnMLab (Ontology Media-Lab)
Ontology Media-Lab (OnMLab) is a next-generation media research lab grounded in ontologies. Guided by the philosophy that “Media is a window to the world,” we strive to build a knowledge ecosystem where data, AI technology, educational content, and domain expertise are woven into a single semantic framework, making specialized knowledge accessible and actionable for everyone. We believe media should transcend simple information delivery to become the starting point for connecting diverse fields and people to create new value.
OnMLab conducts research focused on systematically linking content, data, people, AI, education, industry, and various digital assets through ontologies. These semantic connections ensure that individual pieces of content do not remain isolated information but instead serve as a foundation for the continuous expansion of context and meaning. In this way, media evolves beyond a mere content repository into a living knowledge platform capable of AI-driven understanding and reasoning.
Furthermore, OnMLab pursues a sustainable ecosystem where the value generated through media is not concentrated in specific platforms but is fairly redistributed to all members who produce and connect knowledge. Our ultimate goal is to link the new value created between producers—who provide data, expertise, AI technology, and educational content—and consumers, ensuring that these achievements are shared among all participants in the ecosystem through the medium of our platform.
Through our research, we aim to realize a “future intelligent media ecosystem where humans and AI co-produce knowledge, continuously expanding it through semantic connections and returning that value to benefit all.” We aspire to move beyond the traditional philosophy of media as a mere window reflecting the world, becoming instead a knowledge platform that understands, connects, and grows together with the world.