The psychologist Jung proposed the psychological concept of the collective unconscious, and his theoretical foundation and school are based on this structure. Jung believed that there is a collective subconscious at the bottom of human society. This collective subconscious is shared by all human beings. The collective subconscious does not come from personal direct experience, but from human genetic genes and common consciousness and prototypes created by past collectives. These collective consciousnesses They will mutually influence the future development of individuals and groups, but they may also spread rumors and mistakes, causing mistakes to be passed on repeatedly, affecting the inheritance of knowledge and hindering the development of a civilized society.
This explains the importance of the Decentralized Knowledge Graph (DKG) in verifying the source, copyright, and integrity value of data through blockchain.
Generative AI has been thriving in multiple fields, but it still has many flaws that seriously affect the future development of artificial intelligence across various domains. To prepare generative AI to tackle large-scale social changes, it is necessary to limit the hallucinations, biases, and misjudgments of artificial intelligence, and to eliminate infringements on intellectual property rights.
The decentralized AI graph provides information sources through model output, ensures the verifiability of the presented information, and respects data ownership and sources to address the deficiencies in AI.
The development team of OriginTrail, Trace Labs, has joined the NVIDIA Inception program, hoping to realize a decentralized knowledge graph (DKG) to create a verifiable (Verifiable Internet) AI network.
Trace Labs has implemented the decentralized AI knowledge graph in various fields such as supply chain, healthcare, construction, sports, and aviation, and the collaboration with NVIDIA further integrates blockchain and artificial intelligence perfectly.
How Trace Labs and NVIDIA build a decentralized AI knowledge graph
Origin Trail collaborates with NVIDIA's generative AI to create a 'Decentralized AI Knowledge Graph' based on its own development team's technology.
Retrieval Augmented Generation (RAG) is a mechanism that enhances information retrieval during text generation, providing verifiable and reliable sources of knowledge information. RAG is a technology that allows machine learning models to extract relevant information from external databases before generating output, thereby improving the accuracy of responses and the relevance of context.
Decentralized RAG (dRAG) is an advanced version of RAG, allowing data to exist in the form of Knowledge Assets through OriginTrail's decentralized knowledge graph. Each asset has its specific identification and ownership, ensuring data traceability, integrity, and ownership, significantly improving the accuracy and reliability of GenAI models.
dRAG improves the RAG system by leveraging the decentralized knowledge graph (DKG). Each knowledge asset contains graph data and vector embeddings, invariance proofs, decentralized identifiers (DID), and ownership NFTs. When connected to a permissionless DKG, the structure in the knowledge graph enables a mix of neural networks and symbols with AI, enhancing generative AI models through accuracy input.
Owners of knowledge assets can manage access to the data in the knowledge asset repository, and through blockchain, each piece of knowledge information on the DKG has an encrypted certificate, ensuring that no tampering has occurred since its publication.
NVIDIA Inception and Trace Labs Development Plan
NVIDIA and Trace Labs are developing a decentralized AI knowledge graph through collaboration, providing VC investment opportunities. The Inception program also includes joining the NVIDIA Deep Learning Academy and the NVIDIA Developer Forum, enabling Trace Labs to work with NVIDIA to promote the construction of a decentralized AI ecosystem.
If human society has a collective unconscious, then artificial intelligence also has a collective unconscious of AI, which can redefine the changes that AI can bring to human society.
The application scenario of the decentralized AI knowledge graph is AI agents, which utilize the collective consciousness of the internet at scale to acquire knowledge from shared but sovereign knowledge bases. This means that artificial intelligence can provide coherent and accurate contextual interactions without compromising the privacy and integrity of the data, allowing various professions to establish a trustworthy AI agent ecosystem.
The decentralized AI knowledge graph utilizes NVIDIA's supercomputers to process billions of knowledge assets, laying the scientific foundation for decentralization.
This article discusses how Trace Labs joined NVIDIA's Inception program to collaboratively promote the decentralized AI knowledge graph, which first appeared in Chain News ABMedia.