Research & Writing

Research projects, publications, and articles on technology policy, AI governance, and law.

Writing

2026

Multi-Agent AI is Outpacing the Liability Frameworks Built for Single-Agent Systems — Berkeley Technology Law Journal Blog

Why liability frameworks designed for single-agent AI systems break down when autonomous agents interact, and what multi-agent accountability could look like.

AI Agents are Rewriting the Web's Rules of Engagement. Here's a Way to Fix it. — Tech Policy Press

On how AI agents are disrupting the implicit social contracts of the web, and a proposed framework for agent accountability.

Assistant Neutrality in the Age of Generative AI — Berkeley Technology Law Journal Blog

Examining the concept of neutrality as it applies to AI assistants and the implications for competition and consumer choice.

Research

Cooperation Is Not a Safety Property: A Position on Adversarial-Resistant Agent Identity

July 2026

Position paper presented at the 2nd Technical AI Governance Workshop at ICML 2026, Seoul. Pre-print forthcoming.

AI governance agent identity research

Post-Hoc Explainability and the Transparency Requirements of EU AI Law: An Assessment of Mechanistic Interpretability as an Alternative

June 2026

Presented at CLAIRvoyantS: Workshop on AI Regulation & Safety at ICAIL 2026, Singapore. Camera-ready draft forthcoming.

EU AI law interpretability research

Who Speaks for the Organization? Simulating Strategic Organizational Reasoning in Regulatory Comment Processes

April 2026

Presented at the PoliSim@CHI 2026 workshop (ACM Conference on Human Factors in Computing Systems, Barcelona). Simulating how organizations reason strategically when participating in regulatory comment processes using LLM agents.

AI policy simulation research

No Gate, No Claim: How Existing Law Demands Decentralized Identity for AI Agents

March 2026 – August 2026

Legal analysis across the US, EU, and India demonstrating that existing law already functionally mandates verifiable, decentralized identity for AI agents. $5,000 research grant from the Decentralization Research Center.

AI agents decentralized identity law research

Should We Give AI a Wallet?

February 2026 – May 2026

Research on AI economic rights conducted as part of the Supervised Program for Alignment Research (SPAR), mentored by Larissa Schiavo (Eleos AI), Toni Sims, and Prof. Jeff Sebo (NYU Center for Mind, Ethics, and Policy).

AI alignment economic rights research