E13 E6: The Spatial Web Shift: Building Decentralized AI Knowledge Graphs with Yaniv Tal, Co-Founder & CEO of Geo

Episode Summary

Executive Summary: Yaniv Tal explains Geo as a decentralized, verifiable knowledge network that aims to fix information fragmentation, improve trust, and create a shared factual foundation for discourse and AI. He traces the project’s roots from The Graph and the semantic web, details the multi-layer stack behind Geo, and emphasizes user-first iteration, strong mission alignment, and the importance of building for long-term truth and coordination.

Main Topics: Geo’s mission: trusted, organized knowledge (Priority: 5/5): Geo is positioned as a network for organizing knowledge into a collaborative, provenance-rich knowledge graph so communities can verify information and build shared reality. From The Graph and Web3 to Geo (Priority: 5/5): Yaniv Tal connects Geo to his earlier work on The Graph and the broader Web3 movement, describing it as the realization of long-standing ideas from the semantic web and open data on the web. Building a hybrid protocol + consumer app (Priority: 4/5): Geo is unusual because it combines a core protocol that should be stable with a consumer application that must iterate quickly; that tension shaped product decisions and architecture. Product development, roadmap, and iteration (Priority: 4/5): The team shipped early, learned from users, and made structural changes over time, including starting with native and moving to web for faster iteration after a six-month reset. Scale, architecture, and foundations (Priority: 4/5): Tal argues that for core protocols, getting foundations right matters more than rushing scale, and his earlier experience with The Graph taught hard lessons about indexing and bottlenecks. Future impact: discourse, AI, and coordination (Priority: 5/5): Geo is framed as infrastructure for better discourse, coordination, and AI—serving as a collective memory layer that improves context, accuracy, and alignment. Team, values, and advice to founders (Priority: 3/5): He prioritizes mission alignment, personal drive, and helpfulness in hiring, and advises founders to get products in front of users quickly and ensure they are solving the right problem.

Key Arguments: The internet’s information layer is fragmented and centralized, making it hard to know what is true; Geo addresses this through a verifiable knowledge graph and governance. The original vision of the web was a web of data, not pages, and Geo is an attempt to realize that vision with modern decentralized infrastructure. Core protocols need stability and correctness, while consumer apps need rapid iteration; combining both requires careful product and technical strategy. Early user feedback is essential, but structural changes become harder once a system scales; therefore Geo intentionally delayed scaling to get foundations right. The project’s architecture must include a knowledge graph standard, an on-chain governance layer, indexing/query infrastructure, an SDK, and a browser for attributable browsing. Geo’s long-term value is not just technical; it could improve public discourse, enable better organizational coordination, and support AI with trustworthy context. Hiring should start with mission alignment and personal drive; skills can often be developed if the candidate is deeply motivated and collaborative. Founders should choose problems they can commit to for a decade or more and ship to users as early as possible.

Data Points: Years building Geo: 4 years - Yaniv says Geo has been in active development for the last four years. Time from initial vision: ~9 years - He describes the journey from the 2016–2017 Graph era to Geo as nearly nine years long. Time to first real users: within 6 months - Geo had real users using the product within six months of starting the first version. Iteration cycle length: ~6 months - He says they operated on roughly six-month learning and change cycles. Original start year of The Graph: 2017 - He says he founded The Graph in 2017. Initial version platform: React Native - Geo’s first version was a native app built in React Native before the team restarted on the web. Noise reduction: up to 42 dB - A sponsor readout for Fitnexa Somnipods 3 included this product specification. Weight of earbuds: 3.3 grams - A sponsor readout for Fitnexa Somnipods 3 included this product specification. Thickness of earbuds: under 10 mm - A sponsor readout for Fitnexa Somnipods 3 included this product specification. Web3 ecosystem impact: entire industry brought down several times - Tal recalls The Graph’s indexer outages affecting many Web3 apps during rapid growth.

Pivotal Quotes: "It's a network for organizing knowledge that we can trust." — Yaniv Tal: Defines Geo’s core mission and value proposition. "The job of a core web protocol is to be perfect and to never ever change. And a consumer app's job is to iterate almost hourly, daily, based on user feedback." — Yaniv Tal: Explains the architectural and product challenge of combining protocol stability with app-level speed. "Can we improve the quality of discourse?" — Yaniv Tal: States the north star for Geo’s broader societal impact.

Implications: Geo reflects a broader shift toward verifiable, provenance-based digital infrastructure. If it succeeds, it could reshape how knowledge is organized, how AI gets context, and how institutions and communities reach shared facts and coordination.