Product
Google Model Armor
- First Reported
- Jul 16, 2026
- Latest Reported
- Sep 24, 2026
Reported Context (2)
- Protect AI workloads from prompt injection and jailbreaks using runtime guardrails such as Google Model Armor. Google Cloud and Mandiant outline safeguards for hardening CI/CD pipelines
- as chokepoints, alongside Layer 2 reasoning-based defenses like specialized guard models (such as Model Armor or similar provider-agnostic guardrails) to filter out sensitive data and block malicious prompt Mandiant Outlines Guardrails for AI-Assisted Vulnerability Management
People (13)
MITRE ATT&CK (2)
Vendors (3)
Products (6)
Note: Related entities, including threat actors, malware, CVEs, MITRE ATT&CK techniques, vendors, products, tools, countries, and industries, are shown when they appear in the same reporting. Their presence does not necessarily mean they were targeted, compromised, vulnerable, responsible for the activity, or directly involved in the incident.
