LATIDIA · Robótica
Ataques a la configuración expuesta a la red en robots sociales controlados por LLM
arXiv:2611.22761v1 Tipo de anuncio: cruzado Resumen: Los grandes modelos de lenguaje (LLM) en robots corporizados crean dependencias de seguridad que van más allá de la inferencia. Investigamos un ataque a la configuración expuesta a la red en el robot Pollen
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.22761v1 Announce Type: cross Abstract: Large language models (LLMs) in embodied robots create security dependencies beyond inference. We investigate a network-exposed configuration attack on the Pollen Robotics Reachy Mini Lite, allowing an unauthenticated attacker to replace personality settings and the cloud LLM's system prompt through a management interface, without adversarial conversational input. Eleven prompt-injection attempts across five categories were reported as resisted, whereas both configuration-layer attack paths produced observable behavioural changes on physical hardware. We demonstrate local and network-adjacent access, mapping the weakness to CWE-306 (Missing Authentication for a Critical Function) and CWE-1327 (Binding to an Unrestricted IP Address), with a reported CVSS v3.1 score of 6.5 (Medium). A configuration