LATIDIA · Ciberseguridad

SoK: ¿Agentes de trading o Market Crashers? Disección de fallas de robustez y seguridad en esquemas de trading de maestría en finanzas académicas

arXiv: 2609.19705v1Tipo de anuncio: nuevo Resumen: Los agentes del modelo autónomo de lenguaje grande (LLM) se están moviendo rápidamente hacia dominios de alto riesgo, sin embargo, los estudios de seguridad agentic-AI existentes siguen siendo en gran medida independientes del dominio y overlo

La noticia

arXiv:2609.19705v1 Announce Type: new Abstract: Autonomous large language model (LLM) agents are moving rapidly into high-stakes domains, yet existing agentic-AI security studies remain largely domain-agnostic and overlook the distinctive, high-consequence attack surface such settings create. We examine this gap through financial trading agents, a representative case of high-stakes agentic security, where a single compromised agent has direct execution authority over real capital in an adversarial, reflexive market. To this end, we present FARSIGHT (Financial Agent Robustness and Security Investigation and Global Holistic Testing), a framework that performs scheme-level evaluation of financial LLM agents on two axes: robustness under market turbulence (including flash-crash-like scenarios), and security against three attack types: attacks