Welcome
- Twitter: @johnjnay.
- Fellow at Stanford University - CodeX - Center for Legal Informatics, operated by Stanford Computer Science Dept & Stanford Law School, working on A Legal Informatics Approach to AI Alignment.
- Founder & CEO of Nomos AI.
- Co-founder & Chairman of Brooklyn Investment Group.
- My presentation on Large Language Models & Law to U.S. Congressional staffers, June 2023, can be watched at this link.
- My testimony in front of the Wyoming Legislature on AI Alignment & Law, May 2023, can be watched at this YouTube link.
- My talk on AI Alignment & Law-Informed AI at Stanford FutureLaw, April 2023, can be watched at this YouTube link.
Publications related to AI alignment, and machine learning on legal/regulatory data:
- LegalBench: A collaboratively built benchmark for measuring legal reasoning in Large Language Models
- ARB: Advanced reasoning benchmark for Large Language Models
- Large Language Models as tax attorneys: A case study in legal capabilities emergence
- Large Language Models as fiduciaries: A case study toward robustly communicating with AI through legal standards
- Large Language Models as corporate lobbyists
- Law Informs Code: A legal informatics approach to aligning AI with humans
- Generalizability: AI and humans-in-the-loop
- Predicting law-making with AI
- Gov2Vec: Machine learning distributed representations of government institutions and their legal text
- Natural language processing for legal text
- Predicting human cooperation with AI
- Automatically estimating models of human decision-making with AI
- Natural language processing for presidential legal texts
- Legal informatics
- Big data law
- Hierarchical Bayesian modeling of urban water public policy
- Deriving the market’s political predictions with public policy impact indices constructed with AI
Publications related to AI & climate / markets / impact investing:
- Climate-contingent finance
- Environmental impact bonds: A common framework and looking ahead
- Computational simulation modeling of trader behavior in climate prediction markets
- AI approach to forecasting remotely sensed vegetation health
- Application of AI to prediction of vegetation health
- Decision-support computational models for climate change adaptation
- Empirically-grounded simulation model of impact of climate forecasts on agricultural income
- Agent-based simulation model of the influence of climate forecast accuracy on farmer behavior
- Participatory simulation modeling of urban flooding for decision support
- Modeling agricultural response to drought in the U.S.
- Qualitative and quantitative analysis of drought, risk, and politics
Coverage of some research in the media
- Wyoming Lawmakers Turn To Experts To Stay Ahead Of AI Curve, Even As ChatGPT4 Passes Bar Exam
- Science: Artificial intelligence can predict which congressional bills will pass: Machine learning meets the political machine.
- Vice: GPT can do a corporate lobbyist’s job, study determines
- IEEE Spectrum: AI Goes to K Street: ChatGPT Turns Lobbyist
My (older) open-source software packages (mainly focused on Agent-Based Simulation):
- datafsm: Estimating Decision-making Models with Machine Learning
- predMarket: A Climate Prediction Market Simulation Model
- sa: Sensitivity Analysis for Complex Computational Models
- eat: Empirical Agent Training Software for Data-Driven Modeling
- agp: Agent Gentic Programming Software for Data-Driven Modeling
- forecastVeg: Forecasting Vegetation Health with Machine Learning