Area: Building better digital infrastructure to create a safe and effective rollout of AI across the economy

• Tools for understanding delegation risks and harm mitigation tools for autonomous AI agents.

• Confidentiality, privacy, or verifiability infrastructure for AI and agent deployments.

• Tools for concrete and reliable (and verifiable) supply chain management for AI and data.

• Other infrastructure and governance tools for AI.

Tobin is a researcher at MIT and Stanford, focused on building more robust societal and digital infrastructure for the AI age, with work across privacy, human-centered control, verifiability, and regulatory compliance for AI. Tobin's PhD at MIT was advised by Sandy Pentland at the MIT Media Lab and was fully funded by a Fulbright Future Scholarship. Tobin has a broad research background ranging from confidential and verifiable AI tooling to regulatory contributions to privacy in the future of AI.

Someone who is willing to explore new ideas and collaborate diversely. Mentee will likely have a chance to have a physical presence at Stanford or MIT, and meet with a range of collaborators.