Skip to page content

Toggle Search

Research & Commentary

Featured

Research

Verified Machine Learning Infrastructure: Formal Methods for Trustworthy Artificial Intelligence Deployment
Research

How Does and Should Tort Liability Affect High-Consequence Life Sciences Research?
Expert Insights

Open-Weight AI Models Require Proportional Evaluation Approaches

All RAND CAST Research & Commentary

Skip to results

121 Results

number of items to display at a time

Research

Jul 20, 2026

Restricting AI Agent Use of Biological Tools

Artificial intelligence agents can help non-experts use advanced biological design tools, raising significant misuse concerns. Researchers consider whether safeguards can be built into these tools to block such assistance.
Research

Jul 20, 2026

Securing AI Algorithmic Insights

Algorithmic insights—the novel techniques, methods, and design know-how that materially improve AI systems—are among the most valuable assets in the AI ecosystem. Protecting them from theft or leakage is a matter not only of commercial interest but of national security.
Research

Jul 16, 2026

Pooling Europe's compute: The promise of distributed training for European frontier AI

Frontier AI training has needed compute at one site, hard in Europe given power, chip and political limits. RAND Europe finds distributed training eases the power and political constraints and pools Europe's scattered chips, but adds no new chips.
Expert Insights

Jul 9, 2026

High Bandwidth Memory: What It Is and Why It Matters

High Bandwidth Memory (HBM) is a critical enabling technology for artificial intelligence. The authors of this paper synthesize analysis of the HBM ecosystem for researchers and policymakers.
Research

Jun 26, 2026

Compounding Asymmetries in Nucleic Acid Synthesis Screening: Policy Pathways for Strengthening Global Biosecurity Governance

This perspective argues that narrowing global screening asymmetries requires reframing adoption as a spectrum rather than an all-or-nothing compliance threshold.
Research

Jun 25, 2026

Can LLM Agents Select and Engage with Biological Tools?

Researchers evaluated seven large language model (LLM) agents on their abilities to select and operate biological tools. They find that LLM agents are capable of performing initial interactions with biological tools, which could lower expertise barriers for actors seeking to use them.
Research

Jun 25, 2026

Supply Chain, Energy, and AI Nexus: Evaluating AI Energy Supply Chain Vulnerabilities

The authors identify the inventory of critical electrical equipment for artificial intelligence (AI) data center operations and the impact of supply chain vulnerabilities on the U.S. power grid’s ability to meet AI demand by 2030.
Expert Insights

Jun 15, 2026

Verifiable Audit Trails for AI-Enabled Biological Design Tools: A Proposal for Verifiable Biodesign Logging

The authors propose a tamper-evident audit-trail framework using cryptographic signing to establish a trustworthy chain of custody that can be used with any artificial intelligence–enabled biological design tool.
Research Summary

Jun 10, 2026

What Is Blocking U.S. Power Expansion for AI—and What Could Unlock It by 2030?

Artificial intelligence is increasing electricity demand at a pace that is straining U.S. energy infrastructure. RAND researchers mapped and prioritized barriers to energy expansion in the United States.
Research

Jun 8, 2026

Advancing U.S.–UK Cooperation to Secure Frontier Artificial Intelligence

This interim report identifies a significant opportunity for collaboration between the United States and the United Kingdom and proposes a framework for coordinating efforts to secure frontier artificial intelligence development.
Research

Jun 4, 2026

Verified Machine Learning Infrastructure: Formal Methods for Trustworthy Artificial Intelligence Deployment

The rapid advancement of artificial intelligence systems has created an urgent need to secure the infrastructure on which AI runs. Researchers explored whether formal methods—using mathematical techniques to reason about software behavior—could meaningfully secure AI infrastructure.
Research

May 28, 2026

Investigating the potential use of frontier AI models for offensive cyberattacks: A human uplift study

RAND researchers found that today’s most advanced AI tools can help less-skilled people get started and work faster on cyberattacks, but they do not yet reliably enable them to carry out full, sophisticated attacks.

First
Previous
Page 1 of 11
Next
Last

Other Relevant RAND Research