Home Writing Projects Lab CV

Simulating futures
to build the one we want.

I build open-source software that computes what the law says, measures what it does, and projects what happens next.

Max Ghenis CEO, The Axiom Foundation & PolicyEngine · Washington, DC
Rules what the law says — exact, cited, computable
Impacts what it does — for any household, for whole populations
Projections what happens next — modeled openly, revised as law changes
Work
All projects →
Writing
Four Claude models matched the answer key on 100 AI law questions 95.8% of the time from memory and 99.9% with the rule text I asked four Claude models 100 questions on California's SB 53, New York's RAISE Act, the EU AI Act and US chip rules. From memory, 95.8% of their replies matched the answer key. With the rule text in the prompt, 99.9% did, and with web search, 97.7% did on the 82 questions that version finished. Seven questions account for 39 of the 51 misses from memory. AIs predict that DC banning foie gras would increase the chances of future animal welfare regulations Before DC's November 3 vote, eight AI forecasters, each researching alone, give Initiative 86 an 89.7% chance of passing. After a one-point loss, they give 22% to someone filing a new DC initiative on farmed animals or animal products by the end of 2028; after a one-point win, 31%. Testing California's bond funds against the billionaire tax odds Prediction markets have priced California's Proposition 40 for almost a year, and a fund of California municipal bonds has not moved with the odds by any amount I could detect. Before November 3, I registered a test that pools ballot measures across states. What do forecasters say they don't know? Professional forecasters in the US and euro area have reported full probability distributions for decades. I pooled every one into a tracker and two working papers, and read what they say about growth in the AI era.
All writing →
Research
Enhancing survey microdata with administrative records Microsimulation dataset construction. Presented at NTA 2024. What can LLMs tell us about the ETI? With Jason DeBacker. How language models perceive behavioral responses to tax policy.
All research →