Mihailis Diamantis
University of Iowa, IA·Social Science
Artificial intelligence (AI) can cause real harm, but how should the law respond? Europe's AI Act regulates how businesses build and use AI, and critics worry that this approach may slow innovation.
- Fields
Social Science
Computer Science
- Topics
artificial intelligence
AI regulation
AI governance
- Grants
- 1 known grant, through Sep 2027
What they work on
- Lay the legal foundation for an objective negligence standard that judges AI harms after they occur by comparing AI behavior with an external benchmark.
- Develop AI negligence benchmarks for three representative applications, calibrated to each activity's social value and current safety profile.
- Evaluate the proposed after-the-harm negligence standard against the European Union's AI Act as an alternative model for AI regulation.
More about the research
This project seeks a different, American approach: deciding who is at fault after an AI harm has happened, rather than restricting how AI is developed in advance.
Why it matters
A clear negligence standard for AI harms could give courts, lawmakers and companies a way to assign responsibility when AI causes damage without restricting development in advance. People harmed by AI systems could gain a path to accountability, and developers would have an incentive to build safer software.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Jennifer Mosley
University of Chicago, IL·Social Science
Examine how algorithms that check fidelity to evidence-based practices shape child welfare workers' decisions, by interviewing workers and observing their daily work in partner agencies.
Angel Hsu
University of North Carolina at Chapel Hill, NC·Earth & Environmental Science
Systematically test generative AI models, such as chatbots, to find out how accurately they report information about private-sector climate commitments.
Lu Lin
Pennsylvania State University, PA·Computer Science
Identify critical LLM safety concerns, gaps in current evaluation and infrastructure needs through structured surveys and interviews with experts across computing disciplines.
The school
University of Iowa: 32 labs on PI Pages, $703M a year on research, 418 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.