Insights from a Workshop on Hardware-Enabled Governance Mechanisms

Gabriel Kulp, Margaret Siu, Lennart Heim

Expert InsightsPublished Jun 12, 2025

Hardware-enabled mechanisms (HEMs) are security features in artificial intelligence (AI) chips or related hardware that enact or enable policies on acceptable end-use. To assess HEMs’ potential in bolstering U.S. national security and supporting export controls on advanced AI chips, RAND researchers conducted a virtual workshop in April 2024 with 13 experts from AI and chip industries, civil society, and government. The workshop’s goal was to gather expert perspectives on the potential role of HEMs in AI governance, including their use in preventing unauthorized AI chip use, enforcing export restrictions, and detecting chip smuggling.

The workshop focused on four HEM options and included three activities to gather views on the technical and political feasibility of the four HEMs and possible HEM-based AI governance policies for advancing U.S. national security.

Interviewees shared opinions that no single solution can comprehensively enforce end-use controls, that technically simpler solutions may offer more effective security, and that HEMs could be promising as conditions of international sales and investment deals. During workshop activities, some participants noted that narrow-scope HEMs may be more feasible to develop, whereas broader designs could pose greater security and misuse risks.

These conference proceedings present these and other insights from the workshop and pre-workshop interviews. Though the field has evolved since the workshop concluded, these findings can serve as a reference for researchers exploring ways to develop actionable insights in this field.

Topics

Document Details

Citation

Chicago Manual of Style

Kulp, Gabriel, Margaret Siu, and Lennart Heim, Insights from a Workshop on Hardware-Enabled Governance Mechanisms. Santa Monica, CA: RAND Corporation, 2025. https://www.rand.org/pubs/conf_proceedings/CFA3056-1.html.
BibTeX RIS

Research conducted by

This publication is part of the RAND conference proceeding series. Conference proceedings present a collection of papers delivered at a conference or a summary of the conference.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.