CHAIN-153 — Azure AI Content Safety bypassed via token replay

Summary

Severity: Medium · Likelihood: Medium · Logic: ALL

Why this chain matters

Content Safety screens LLM outputs, but the moderation verdict is cached client-side for replay. An attacker captures a 'safe' verdict and replays it for different content, bypassing moderation.

Component rules

This chain fires when its trigger conditions are met by the following rules. Click any rule to see its detection logic and compliance mappings.

Rule ID	Role
zt_ai_006	Trigger
zt_vis_002	Trigger

Attack walkthrough

Step 1 — Observe moderation verdict traffic; capture 'safe' signatures.

Actor: Attacker
MITRE ATT&CK: T1557.002
Enabled by: zt_ai_006

Attacker gain: Captured 'safe' moderation token.

Step 2 — Submit harmful content, present cached token; app accepts.

Actor: Attacker
MITRE ATT&CK: T1550.004
Enabled by: zt_vis_002

Attacker gain: Moderation bypass.

Blast radius

Initial access	Client-side moderation cache.
Max privilege	Moderation bypass.
Data at risk	Moderation integrity
Services at risk	Content Safety consumers

How the logic works

The chain fires only when every rule above has at least one finding in the current scan. Missing any one rule breaks the chain — so remediating any single step disrupts the attack path.