3 papers
cs.CY2026
AI Security Priorities: A Field-Wide Agenda
Gil Gekker, Rachel Steratore, Everett Smith +9
As AI systems are rapidly integrated into critical economic, governmental, and national security functions, the gap between AI adoption and AI security readiness continues to widen…
cs.CR2026
Practice-Informed, Practice-Ready: An AI security incident taxonomy
Lukas Bieringer, Sean McGregor, Nicole Nichols +8
With the increasing prevalence of AI systems, several real-world AI security incidents have been reported. However, despite forthcoming legal mandates, the reporting and collection…
cs.CR2026
Proactive Hardening of LLM Defenses with HASTE
Henry Chen, Victor Aranda, Samarth Keshari +2
Prompt-based attack techniques are one of the primary challenges in securely deploying and protecting LLM-based AI systems. LLM inputs are an unbounded, unstructured space. Consequ…