MLOps.community

Collective Memory for AI on Decentralized Knowledge Graph // Tomaž Levak // #285

Jan 24, 2025
Tomaž Levak, Co-founder and CEO of Trace Labs, dives into the world of decentralized knowledge graphs and their role in AI. He discusses how these graphs enhance data integrity and privacy while promoting collaboration among organizations. Practical use cases in enterprise sectors are highlighted, showcasing their economic potential. Levak also explores the fusion of AI and personal health management, emphasizing innovative technologies that improve well-being. The conversation concludes with insights on the future of decentralized AI and its convergence with blockchain.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Decentralized Knowledge Graph for AI

  • OriginTrail builds a decentralized knowledge graph (DKG) for symbolic AI.
  • Decentralization improves data ownership, verifiability, and transparency.
INSIGHT

Paranets within the DKG

  • The DKG is structured as a "knowledge graph of knowledge graphs" with specialized "paranets" or parallel networks.
  • Paranets allow for domain-specific rules, data structures, and ontologies within the larger DKG.
ANECDOTE

Enterprise Use Cases of DKG

  • Fortune 500 companies and Swiss Federal Railways use the DKG with private paranets for secure data exchange.
  • Importers use the DKG to exchange audit reports with Homeland Security, demonstrating a specialized use case.
Get the Snipd Podcast app to discover more snips from this episode
Get the app