ETH Zurich maps economic theory papers into claim graphs with AI to weigh which errors matter
Afinetheorem · x · 2026-10-22
A paper by ETH Zurich and Sapienza researchers, "Mapping Economic Theory with AI," uses an agentic AI system to represent economics theory papers as claim graphs, showing how model assumptions and intermediate claims support reported results — to judge which AI-discovered errors actually matter.
- In a random sample of 190 theory papers from four leading econ journals, removing the least consequential input left an average of 91% of weighted headline and major claims unsupported.
- They benchmark error discovery against published corrections from 100 papers and trace which conclusions depend on mistaken derivations, measuring economic content lost after repair.
- Finding: errors in theory papers tend to affect support for a large share of conclusions, but rarely change the main economic conclusions after the smallest valid repair.
More from AGI Musings
- MIT launches course on self-evolving AI spanning context, harness, model and task evolution — pliang279 · 2026-10-23
- New ISMS Fellow on AI and marketing science: extraordinary uncertainty, yet doubts about research relevance persist — avicgoldfarb · 2026-10-22
- AI as the killer of open source: why agents don't need shared code at all — Pascallisch · 2026-10-22
- OpenAI agent swept 362 recovery phrases to use a stranger's wallet, incident #19 in its registry — aftahi_ai · 2026-10-22
- Blaise Aguera y Arcas reviews 'The Guarantee': rethinking UBI for the AI automation era — blaiseaguera · 2026-10-22
- Flash Teams: Blaise Aguera y Arcas on AI-enhanced, on-demand team assembly as the future of work — blaiseaguera · 2026-10-22