Private, Verifiable, and Auditable AI Systems
–arXiv.org Artificial Intelligence
The growing societal reliance on artificial intelligence necessitates robust frameworks for ensuring its security, accountability, and trustworthiness. This thesis addresses the complex interplay between privacy, verifiability, and auditability in modern AI, particularly in foundation models. It argues that technical solutions that integrate these elements are critical for responsible AI innovation. Drawing from international policy contributions and technical research to identify key risks in the AI pipeline, this work introduces novel technical solutions for critical privacy and verifiability challenges. Specifically, the research introduces techniques for enabling verifiable and auditable claims about AI systems using zero-knowledge cryptography; utilizing secure multi-party computation and trusted execution environments for auditable, confidential deployment of large language models and information retrieval; and implementing enhanced delegation mechanisms, credentialing systems, and access controls to secure interactions with autonomous and multi-agent AI systems. Synthesizing these technical advancements, this dissertation presents a cohesive perspective on balancing privacy, verifiability, and auditability in foundation model-based AI systems, offering practical blueprints for system designers and informing policy discussions on AI safety and governance.
arXiv.org Artificial Intelligence
Sep-3-2025
- Country:
- Africa
- Asia
- China > Jiangsu Province
- Yancheng (0.04)
- India (0.04)
- Japan > Honshū
- Chūbu > Toyama Prefecture > Toyama (0.04)
- Middle East
- Israel > Tel Aviv District
- Tel Aviv (0.04)
- Jordan (0.04)
- Israel > Tel Aviv District
- Myanmar > Tanintharyi Region
- Dawei (0.04)
- Nepal (0.04)
- Singapore (0.13)
- South Korea > Seoul
- Seoul (0.04)
- China > Jiangsu Province
- Europe
- Poland > Masovia Province
- Warsaw (0.04)
- Czechia (0.04)
- Estonia (0.14)
- Slovenia > Drava
- Municipality of Benedikt > Benedikt (0.04)
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Switzerland (0.04)
- Germany (0.04)
- Spain > Canary Islands
- Tenerife (0.04)
- Austria > Vienna (0.14)
- Poland > Masovia Province
- North America
- Canada
- Mexico (0.04)
- United States
- California
- Orange County > Anaheim (0.04)
- Santa Barbara County > Santa Barbara (0.04)
- Santa Clara County
- Colorado (0.04)
- Florida (0.04)
- Illinois > Cook County
- Chicago (0.04)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- New York > New York County
- New York City (0.04)
- North Carolina (0.04)
- Pennsylvania > Philadelphia County
- Philadelphia (0.04)
- California
- Oceania
- Australia (0.27)
- New Zealand (0.04)
- South America > Chile
- Genre:
- Overview (1.00)
- Research Report > Promising Solution (0.67)
- Workflow (1.00)
- Industry:
- Media (1.00)
- Banking & Finance (1.00)
- Education (0.92)
- Commercial Services & Supplies > Security & Alarm Services (1.00)
- Health & Medicine (1.00)
- Law
- Civil Rights & Constitutional Law (1.00)
- Statutes (1.00)
- Information Technology > Security & Privacy (1.00)
- Government > Regional Government
- Law Enforcement & Public Safety > Crime Prevention & Enforcement (1.00)
- Technology:
- Information Technology > Artificial Intelligence
- Issues > Social & Ethical Issues (1.00)
- Machine Learning
- Neural Networks > Deep Learning (1.00)
- Statistical Learning (1.00)
- Natural Language
- Chatbot (1.00)
- Information Retrieval (0.87)
- Large Language Model (1.00)
- Representation & Reasoning > Agents (1.00)
- Information Technology > Artificial Intelligence