Stanford University — LLM-Generated Research Ideation Benchmark

[Navigating Transformative AI] Open Philanthropy recommended a grant of $880,000 over two years to Stanford University to support a project evaluating the abilities of Large Language Model (LLM) agents at generating Machine Learning (ML) research project ideas via a large-scale experiment, led by Assistant Professor Tatsunori Hashimoto. This grant was funded via a request for proposals for projects benchmarking LLM agents on consequential real-world tasks and falls within our focus area of potential risks from advanced artificial intelligence.