By: Gab
Subnet 62 – Ridges: AI-Powered Research Engine
Ridges is designed to organize and rank the world’s research knowledge.
Problem Solved:
The internet’s research content is fragmented, unverified, and buried under irrelevant search results.
How It Works:
Miners provide models that ingest academic papers, research datasets, and technical knowledge. Ridges scores them based on accuracy, novelty, and usefulness.
Why It’s Valuable:
➡️ Builds a trustworthy research layer for AI models.
➡️ Can power specialized search for scientists, biotech, and enterprise R&D.
➡️ Data advantage compounds over time, the more knowledge it processes, the better it gets.
Big Picture:
Ridges could become the “Google Scholar 2.0” of decentralized AI, which is highly defensible and hard for competitors to replicate.
Subnet 41 – Sportensor: Real-Time Sports Intelligence
Sportensor focuses on real-time sports predictions, analytics, and content generation.
Problem Solved:
Sports bettors, fantasy leagues, and broadcasters need instant, accurate insights not delayed or biased by centralized providers.
How It Works:
Miners provide prediction models for live games (odds, performance stats, play outcomes) across multiple sports. Validators reward models that prove accurate in real-time.
Why It’s Valuable:
➡️ Massive addressable marketsports betting and analytics exceed $300B+ globally.
➡️ Delivers a trustless prediction oracle for web3 gaming, fantasy sports, and decentralized sportsbooks.
➡️ Community-driven model updates keep it ahead of centralized APIs.
Big Picture: Sportensor can dominate the sports data + AI prediction market, a space where live accuracy and decentralization are key.
Why Both Will Lead
➡️ They solve real-world, high-demand problems with massive existing markets.
➡️ They’re data-compounding subnets the more they’re used, the stronger their moat.
➡️ Their outputs can feed into other subnets (like trading, news summarization, and language models), making them core infrastructure.
