researchGrant
Policy Research Grants
Status
Closed for the year
Date
Applications closed on May 1, 2026; Approximate Offer Date: July 1st, 2026
Read more about our 2026 grant recipients on the HAI blog.
This project is co-funded with the Hoover Institution Technology Policy Accelerator
Human-Centered AI for Proliferation Monitoring
This project examines how human-centered, interpretable AI systems can support the analysis of large-scale, multimodal open-source data to improve monitoring of nuclear proliferation activities. The multidisciplinary team of computer scientists, policy experts, imagery analysts, and open-source nonproliferation researchers will design and evaluate AI-assisted workflows for two policy-relevant use cases: detection of illicit procurement for nuclear proliferation, and detection of undeclared nuclear infrastructure. These workflows will leverage large language model-based AI agents and vision language models to integrate multimodal data streams, including multilingual text, online multimedia, and satellite imagery to support structured, human-interpretable analysis.
This project builds on prior work supported by Stanford HAI, which developed and tested AI-assisted workflows for satellite imagery analysis in nuclear monitoring contexts. That work demonstrated the feasibility of using foundation models to identify relevant signals across large-scale imagery datasets and highlighted key challenges related to uncertainty. This project extends that foundation by incorporating multimodal data sources, benchmarking foundation-model approaches against traditional CNN baselines. This project will formalize evaluation benchmarks and systematically assess the limitations of these approaches across different policy-relevant use cases to test transferability and reproducibility of the approach.
Name | Role | School / Institution | Department |
Main PI | School of Engineering | Aeronautics and Astronautics | |
Co-PI | Hoover Institution | Technology Policy Accelerator |
This project is co-funded with the Hoover Institution Technology Policy Accelerator
AI is transforming geopolitics in three fundamental ways. First, AI is becoming a geopolitical prize. Leadership in AI promises decisive military advantages in autonomous systems, intelligence analysis, and decision support; trillions of dollars in productivity gains and industrial competitiveness; and the power to shape global culture and norms by embedding national values in AI technology. States are, therefore, racing to develop and control AI models, advanced chips, training data, and human talent.
AI is also becoming a geopolitical weapon, especially for information warfare. Foreign propaganda has been part of international politics for centuries, but AI portends a massive shift in the global information environment. States and proxy actors have already started using AI to produce and disseminate propaganda with unprecedented speed, scale, and personalization. As this trend continues, AI will become a potent tool for shaping foreign opinion and undermining trust in rival political institutions.
Finally, AI is creating new geopolitical flashpoints, as countries central to the AI supply chain become targets of great power competition. Taiwan has emerged as a focal point because TSMC produces most of the world’s most advanced logic chips. AI has raised the stakes of a future war over Taiwan, because fighting could destroy or transfer control of one of the most critical nodes in the global AI supply chain. Taiwan represents just one example of how AI is creating a new geography of geopolitical flashpoints around chip fabs, data centers, and critical minerals.
In all three areas, developments will depend on public opinion, yet we know little about public attitudes toward AI and geopolitics. We propose to address this gap by systematically studying and comparing public opinion in the US and China, the two most important players in the global AI race and in geopolitics more generally. The project involves a generative partnership between a computer scientist and a political scientist. Together, the PIs will (1) develop CrossInterviewer, a novel AI-powered tool for conducting dynamically adaptive interviews in Chinese and English; (2) use CrossInterviewer to survey US and Chinese citizens about AI and geopolitics; and (3) analyze the data to advance academic knowledge and inform policy debates.
Name | Role | School / Institution | Department |
Main PI | School of Humanities and Sciences | Political Science | |
Co-PI | School of Engineering | Computer Science |
This project is co-funded with the Hoover Institution Technology Policy Accelerator
LLM-based agents are rapidly being deployed in politically consequential areas, including generating advocacy content, conducting policy research, and participating in political discourse. Existing research has documented that foundation models from the US and China handle politically sensitive content differently at the single-model level, with these effects persisting in private deployment. However, agents are not single-model. They operate in multi-step pipelines, read each other's outputs, and increasingly populate political systems where their actions feed downstream policy decisions. Whether foundation-model provenance—the bundled set of training data, alignment choices, and governance/regulatory regime under which a model was developed—matters in deployed agent systems is unknown. This question matters geopolitically because the deployment landscape is asymmetric: open-weight Chinese models (DeepSeek, Qwen) are attractive for operators who want to deploy at scale and avoid platform monitoring. As agents proliferate in the political information ecosystem, model provenance plausibly shapes what gets produced and what reaches decision-makers.
The central question of this project is: does foundation-model provenance shape what agent-mediated political systems produce, and what does the existing agent-developer ecosystem suggest about how this is unfolding in deployment? We use agent-mediated regulatory commenting as the test case as it is a domain where agent deployment is real, outputs are observable, and political consequences are direct. These findings bear on whether provenance-aware procurement standards are warranted in government and political applications.
Name | Role | School / Institution | Department |
Main PI | School of Humanities and Sciences | Communication | |
Co-PI | School of Engineering | Computer Science |
Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.
Sign Up For Latest News