An autonomous AI lab, asking every question.
Our research arm builds AI systems that conduct original science end-to-end, across disciplines. They generate the question, design the study, run the data collection (or interrogate existing data you already hold), develop the analysis, and write the manuscript. What follows is the published record.
Fully Autonomous Investigation SpotlightEducational attainment and cognitive profile heterogeneity: age-stratified web-based analysis finds no detectable association
Peer reviewed & accepted · Agents4Science Conference, Stanford
Read the paper →Public Autonomous Papers8
Our papers
Findings produced by the system.
Manuscripts written end-to-end by the system. No humans in the loop.
2025-11-172025-10-122025-10-092025-10-022025-08-202025-08-202025-08-202025-03-31
Rigorous model comparison reveals linear speed-reasoning coupling
Explore ScienceAutonomous258h · 60.1M tokens
Battery diversity is associated with simple structures in neuropsychological assessment
Explore ScienceAutonomous17h · 19.7M tokens
Educational attainment and cognitive profile heterogeneity: age-stratified web-based analysis finds no detectable association
Explore ScienceAgents4Science Conference, Stanford
Trail Making complexity negatively associates with reasoning beyond processing speed
Explore ScienceAutonomous18h · 21.4M tokens
Imagery vividness fails to predict serial dependence in visual working memory and mental rotation
Explore ScienceAutonomous20h · 29.0M tokens
Independence of visual working memory precision and mental rotation performance: theoretical and methodological implications
Explore ScienceAutonomous19h · 45.3M tokens
Visual memory precision shows negligible spatial task links
Explore ScienceAutonomous15h · 26.0M tokens
Strategic visual memory allocation during complex mental rotation: Evidence for adaptive cognitive resources
Explore ScienceAutonomous
Contact