Top AI Product

SciAtlas, from UCL, is a large-scale knowledge graph aimed at automating scientific research — structuring the relationships across papers, methods, datasets, and findings so an AI system can navigate the literature the way a domain expert does.

## The problem

Scientific literature expands faster than any human can track. Automated knowledge-graph construction is a hot 2026 research area, but the hard parts are recognizing long multi-word scientific entities, generalizing across domains, and capturing the hierarchical structure of knowledge. A knowledge graph that handles these well becomes the substrate an AI research agent can reason over.

## Why it matters

The “AI for science” push — from OpenAI’s recent Erdős-conjecture result to autonomous research agents — needs structured knowledge to reason against, not just raw text. SciAtlas is infrastructure for that: the map an automated researcher uses to find gaps, connect distant findings, and propose experiments. As research agents get more capable, the bottleneck shifts from reasoning to having a reliable, queryable representation of what’s already known. SciAtlas is a bet on building that representation at scale.