Selected work

Representative Papers

Designing Incident Reporting Systems for Harms from General-Purpose AI

Nov 08, 2025

In response to escalating safety and rights risks posed by general-purpose artificial intelligence (GPAI), this paper proposes the first systematic reporting framework for GPAI incidents. Drawing on a systematic literature review and cross-case analysis of high-stakes domains—including aviation and healthcare—as well as regulatory practices in the U.S. and EU, the study identifies seven core dimensions: policy objectives, reporting entities, incident typologies, reporting modalities (mandatory vs. voluntary), near-miss inclusion, anonymity safeguards, and legal immunity provisions. It critically examines the trade-offs among safety learning, cross-organizational information sharing, and legal interoperability inherent in each mechanism. The resulting framework offers policymakers and researchers an actionable, theory-informed blueprint for designing GPAI incident reporting infrastructure—addressing a critical gap in GPAI risk governance and advancing the institutional foundations for responsible AI development and deployment.

1 citationsRead paper

Characterizing Agentic Flooding of Government Services

Aug 17, 2026

This study addresses the systemic overload of government services caused by AI agent-induced flooding. Pioneering the concept of "agent flooding," this research constructs a risk matrix to assess exposure levels and conducts empirical analysis integrating multi-jurisdictional case studies with policy mapping. The findings reveal current flooding patterns, identify high-risk services, and propose mitigation strategies that ensure equity without performance trade-offs, alongside actionable short-term recommendations. By bridging the governance gap in AI-mediated public service delivery, this work establishes a systematic framework combining theoretical innovation with practical guidance for managing emerging digital risks. Ultimately, it provides critical support for enhancing the resilience of government infrastructure against novel technological disruptions.

0 citationsRead paper

Demonstrating Restraint

Feb 20, 2026

This study addresses the strategic risks posed by unilateral U.S. advancement in artificial intelligence, which may be perceived by adversaries as an existential threat, potentially triggering preemptive and destabilizing responses that jeopardize national security. Drawing on the theory of credible commitments from international relations, the paper proposes a framework for establishing multilayered mechanisms of credible restraint through coordinated policy and technical measures. These mechanisms aim to enhance the credibility of commitments under conditions of high uncertainty. The analysis demonstrates that, compared to pursuing unilateral advantage, a strategy of credible restraint more effectively mitigates adversaries’ misperceptions regarding both capabilities and intentions, thereby improving strategic stability. This approach offers a novel pathway for international security governance in the AI era.

0 citationsRead paper

What do model reports say about their ChemBio benchmark evaluations? Comparing recent releases to the STREAM framework

Oct 23, 2025

This study addresses the lack of methodological transparency in chemical and biological (ChemBio) misuse risk assessment reports issued by leading AI developers—OpenAI, Anthropic, and Google DeepMind—in 2025. It conducts the first cross-organizational, systematic comparison grounded in the STREAM v1 framework. Using qualitative analysis and structured scoring across key dimensions—including test examples, prompt engineering conditions, and controllable elicitation strategies—the study reveals that all three reports omit concrete test materials and precise, reproducible triggering conditions. Its primary contributions are threefold: (1) introducing the first standardized transparency evaluation paradigm specifically designed for ChemBio risk assessments; (2) identifying shared methodological shortcomings across institutions; and (3) proposing an actionable, integrative pathway toward cross-organizational best practices. Collectively, these advances establish a foundational methodology to enhance scientific rigor, reproducibility, and industry alignment in AI safety evaluation.

0 citationsRead paper

STREAM (ChemBio): A Standard for Transparently Reporting Evaluations in AI Model Reports

Aug 13, 2025

Current AI model evaluations in chemical and biological (ChemBio) safety suffer from opaque reporting and a lack of standardized disclosure practices. Method: This paper introduces the first transparent reporting standard specifically for ChemBio risk assessment of AI models. Drawing on best practices from government, academia, and industry, we develop a structured reporting framework, a standardized evaluation metadata schema, a concise three-page operational report template, and multiple “gold-standard” exemplar reports. Contribution/Results: We systematically define, for the first time, the disclosure dimensions and quality requirements for ChemBio safety evaluations; significantly improve the completeness and reproducibility of assessment information; and enable third-party independent auditing and cross-model comparability. The standard has been adopted by multiple leading AI research organizations, enhancing both public trust and methodological rigor in ChemBio safety assessment.

0 citationsRead paper

Recent publications

Latest Papers

Characterizing Agentic Flooding of Government Services

Aug 17, 2026

This study addresses the systemic overload of government services caused by AI agent-induced flooding. Pioneering the concept of "agent flooding," this research constructs a risk matrix to assess exposure levels and conducts empirical analysis integrating multi-jurisdictional case studies with policy mapping. The findings reveal current flooding patterns, identify high-risk services, and propose mitigation strategies that ensure equity without performance trade-offs, alongside actionable short-term recommendations. By bridging the governance gap in AI-mediated public service delivery, this work establishes a systematic framework combining theoretical innovation with practical guidance for managing emerging digital risks. Ultimately, it provides critical support for enhancing the resilience of government infrastructure against novel technological disruptions.

0 citationsRead paper

Demonstrating Restraint

Feb 20, 2026

This study addresses the strategic risks posed by unilateral U.S. advancement in artificial intelligence, which may be perceived by adversaries as an existential threat, potentially triggering preemptive and destabilizing responses that jeopardize national security. Drawing on the theory of credible commitments from international relations, the paper proposes a framework for establishing multilayered mechanisms of credible restraint through coordinated policy and technical measures. These mechanisms aim to enhance the credibility of commitments under conditions of high uncertainty. The analysis demonstrates that, compared to pursuing unilateral advantage, a strategy of credible restraint more effectively mitigates adversaries’ misperceptions regarding both capabilities and intentions, thereby improving strategic stability. This approach offers a novel pathway for international security governance in the AI era.

0 citationsRead paper

Designing Incident Reporting Systems for Harms from General-Purpose AI

Nov 08, 2025

In response to escalating safety and rights risks posed by general-purpose artificial intelligence (GPAI), this paper proposes the first systematic reporting framework for GPAI incidents. Drawing on a systematic literature review and cross-case analysis of high-stakes domains—including aviation and healthcare—as well as regulatory practices in the U.S. and EU, the study identifies seven core dimensions: policy objectives, reporting entities, incident typologies, reporting modalities (mandatory vs. voluntary), near-miss inclusion, anonymity safeguards, and legal immunity provisions. It critically examines the trade-offs among safety learning, cross-organizational information sharing, and legal interoperability inherent in each mechanism. The resulting framework offers policymakers and researchers an actionable, theory-informed blueprint for designing GPAI incident reporting infrastructure—addressing a critical gap in GPAI risk governance and advancing the institutional foundations for responsible AI development and deployment.

1 citationsRead paper

What do model reports say about their ChemBio benchmark evaluations? Comparing recent releases to the STREAM framework

Oct 23, 2025

This study addresses the lack of methodological transparency in chemical and biological (ChemBio) misuse risk assessment reports issued by leading AI developers—OpenAI, Anthropic, and Google DeepMind—in 2025. It conducts the first cross-organizational, systematic comparison grounded in the STREAM v1 framework. Using qualitative analysis and structured scoring across key dimensions—including test examples, prompt engineering conditions, and controllable elicitation strategies—the study reveals that all three reports omit concrete test materials and precise, reproducible triggering conditions. Its primary contributions are threefold: (1) introducing the first standardized transparency evaluation paradigm specifically designed for ChemBio risk assessments; (2) identifying shared methodological shortcomings across institutions; and (3) proposing an actionable, integrative pathway toward cross-organizational best practices. Collectively, these advances establish a foundational methodology to enhance scientific rigor, reproducibility, and industry alignment in AI safety evaluation.

0 citationsRead paper

STREAM (ChemBio): A Standard for Transparently Reporting Evaluations in AI Model Reports

Aug 13, 2025

Current AI model evaluations in chemical and biological (ChemBio) safety suffer from opaque reporting and a lack of standardized disclosure practices. Method: This paper introduces the first transparent reporting standard specifically for ChemBio risk assessment of AI models. Drawing on best practices from government, academia, and industry, we develop a structured reporting framework, a standardized evaluation metadata schema, a concise three-page operational report template, and multiple “gold-standard” exemplar reports. Contribution/Results: We systematically define, for the first time, the disclosure dimensions and quality requirements for ChemBio safety evaluations; significantly improve the completeness and reproducibility of assessment information; and enable third-party independent auditing and cross-model comparability. The standard has been adopted by multiple leading AI research organizations, enhancing both public trust and methodological rigor in ChemBio safety assessment.

0 citationsRead paper