Policy Explorer

Cosmos Institute granted Caleb Maresca on August 15, 2025 for Policy Explorer, with funding in the $1K to $10K range plus compute credits. The grant came through the inaugural AI x Truth-Seeking cohort of 27 winners funded jointly by Cosmos Institute and FIRE under a $1M program.

Maresca is a PhD student at New York University whose research sits at the intersection of AI and economic systems, including macroeconomic modeling of the effects of AI, lab experiments evaluating the strategic behavior of AI agents, ML for causal inference, and mechanistic interpretability. His preprint "Strategic Wealth Accumulation Under Transformative AI Expectations" (arXiv 2502.11264) argues that the anticipation of advanced AI alone can drive substantial economic adjustment ahead of any technological breakthrough. He also developed SenGen, a Python package for generating interactive scenarios for training and evaluating LLM-based reinforcement learning agents.

Policy Explorer is associated with the explorepolicy.org project, which positions itself as a causal AI system for policymakers to visualize causal relationships and simulate interventions before implementation. Listed scenarios include emergency response optimization, workforce retraining, grid balancing, and healthcare resource allocation. The tool's design assumption is that policy work, like science and engineering, should be exploratory: capable of simulating, testing, and adapting under uncertain technological conditions.

Within the cohort Policy Explorer sits in the agents and governance layers. It is distinguished from other policy-adjacent grantees by focusing on assumption-surfacing and causal simulation rather than drafting or summarization, which puts it closer to econometric simulation tools than to LLM policy-writing assistants.