The Trust Problem in Onchain Insights: What Data Can We Believe?
summary
Claims that raw onchain state is the single source of truth while dashboards and marketing narratives introduce noise; tools used are The Graph’s subgraphs to substreams for indexing, Dune/DeFiLlama/CoinGecko for ad hoc checks, and Flipside’s chat + AI that exposes provenance and procedures. Concrete result: Flipside scores airdrop addresses with wallet health scores 0–10, shifting targets from 2–3 (pump‑and‑dump) to 7–8 (engaged users). AI can decode transaction hashes and cite tables but needs expert‑curated training and faster indexing to be fully reliable.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:09Trustworthy blockchain insights
- 7:23Verifying data sources
- 9:12Onchain data as truth
- 11:25AI analyzing blockchain
- 16:40Flipside Crypto AI access
- 20:32Confidence scores for AI
- 22:21Indexing in blockchain
- 24:29Subgraphs and substreams
- 26:36Decentralized APIs
- 29:50Standardizing queries
- 30:33Verifying blockchain data
related talks
37mPanel: The Hidden Value behind Cryptoanalytics
25mYou Are What You Ingest: Why Data Quality Matters to Smart Contracts
49mCrypto FinTech Systems: Can the Ultimate Trust Machine Be Trusted?
28mBeyond the Balance Sheet: Transactional Data as a Window to Company Health
49mBlockchains for Artificial Intelligence
17mDefi the future of finance
13mDream of Transparent and Secure CEX
23mHow to overcome AI threat to user authenticity