Can /ai audit science? 🤔

Matt Schlicht and @yesnoerror are building an /ai agent framework to fact-check and verify scientific research.

Their goal? Ensure accuracy in published studies by detecting errors and inconsistencies using advanced /ai models.

For now, they’re laser-focused on perfecting this mission. But in the future? Open-source tools, APIs, and expanding beyond science could be next. 🚀

This is an experiment! yesnoerror is an AI agent that uses OpenAI's o1 model to find errors in published research papers