Somesh Jha

Secure-CUA: Controlling Untrusted Influence in Computer-Use Agents

Beyond Predictable Paths: Redefining AI Security Incident Reporting for Agents

Differentially Private Paired Table-Image Multimodal Synthesis

Rising From the Ashes: How Agentic AI is Unblocking Challenges in Cybersecurity

Efficient and Sound Probabilistic Verification for AI Agents

Confused ChatGPT: Cross-App Context Poisoning via First-Party APIs

Agent Security is a Systems Problem

Undetectable Backdoors in Model Parameters: Hiding Sparse Secrets in High Dimensions

Dependency-Aware Privacy for Multi-turn LLM Agents

PolicyBank: Evolving Policy Understanding for LLM Agents

Formal Policy Enforcement for Agentic Systems

Auto-SPT: Automating Semantic Preserving Transformations for Code

Systems Security Foundations for Agentic Computing