logo

Subverting AIOps Systems Through Poisoned Input Data

ID: ff892c86-bbcd-5bec-9d8d-c475b29df494

STIX ID: report--ff892c86-bbcd-5bec-9d8d-c475b29df494

Feed Name: Schneier on Security

Threat Score
60/100

Date Published: 2025-08-20

Date Updated: 2026-04-19

Author: Bruce Schneier

...
...

This summary describes a 2025 academic paper showing that LLM-driven AIOps systems can be subverted via poisoned telemetry data: researchers present an automated attack pipeline (AIOpsDoom) that injects adversarial telemetry to induce harmful remedial actions, and propose a defense (AIOpsShield) to sanitize telemetry and mitigate such attacks—highlighting AIOps as an emerging attack surface with potential for infrastructure compromise.

Your team is not currently subscribed to this feed. You must subscribe to it in order to see this post.