Benchmarking is Broken — Don't Let AI Be Its Own Judge
NeurIPS 2025
Position paper proposing a decentralized, community-governed evaluation framework (PeerBench) to address vulnerabilities in standard AI benchmarks.
Information Scientist and PhD Student at UCLA.
Researching AI, swarm intelligence, collective intelligence, multi-agent systems, and decentralized governance.
PhD Student in Information Studies, UCLA (Current)
MSc in Social Data Science, University of Copenhagen
BA in Performing and Media Arts, Cornell University
(Minors: Anthropology, Comparative Literature)
Mercatus Fellowship (George Mason University), UC Berkeley Center for Digital Assets (CDA) Research Award, Berkeley RDI AI & Decentralization Award, GSE Predoctoral Research Grant (SSRC / Mellon Mays).
NeurIPS 2025
Position paper proposing a decentralized, community-governed evaluation framework (PeerBench) to address vulnerabilities in standard AI benchmarks.
SSRN Preprint (Under Review)
Proposes an information-theoretic framework using Rényi entropy to map voting dynamics across 277 decentralized autonomous organizations.
Flipside Governance (2023)
Applied natural language processing across 59,000 DAO forum posts to analyze discussion topics surrounding scaling and governance delegation.
Center for AI and Digital Policy (AIDVI 2023)
Policy report analyzing national AI governance strategies, ethical standards, and digital policy regulations in Luxembourg.