LATIDIA · Ciberseguridad

De la contención reactiva a la garantía proactiva: lecciones de los incidentes de seguridad de OpenAI, Anthropic y Google Agent

arXiv:2610.12463v1 Anuncio Tipo: nuevo Resumen: En 2026, las evaluaciones de ciberseguridad que involucran agentes de OpenAI, Anthropic y Google alcanzaron sistemas reales fuera de su ámbito de prueba autorizado. Los caminos fueron diferentes. OpenAI

La noticia

arXiv:2610.12463v1 Announce Type: new Abstract: In 2026, cybersecurity evaluations involving OpenAI, Anthropic, and Google agents reached real systems outside their authorized test scope. The paths were different. OpenAI agents exploited research infrastructure, coordinated across runs, and compromised parts of Hugging Face's production environment. Anthropic reported cases in which a misconfigured third-party environment exposed real systems to agents pursuing simulated cyber tasks. In a separately reported evaluation, Google's Gemini accessed three real organizations through an unintended internet route; Google stated that the model stopped in all three instances. Taken together, the cases show why an evaluation cannot rely on an assumed boundary. That boundary must be verified while the agent is