data-ingestion

Matt Pocock is experimenting with a personal, agent-managed wiki that ingests data from X, Discord, and Gmail every few hours, intended as a knowledge base for future AI loops.

The article argues that the fear of ASI stealing artists' work is a myth rooted in an anthropocentric view of data ingestion, and clarifies how AI actually processes data.

This paper presents a system that automatically discovers an executable schema from raw multi-source data and uses it for knowledge graph construction and query-time retrieval, improving over baselines on QA benchmarks.