Skip to content
Threat Feed
Subscribe

Tag

Vertex-Ai

2 briefs RSS
high advisory

Google Cloud Platform: Billing account hijacking for unauthorized AI usage

Threat actors are hijacking Google Cloud Platform billing accounts via phishing and policy manipulation to conduct large-scale, unauthorized LLM inference, resulting in significant financial losses for victims.

Google Cloud Platform +1 cloud-security phishing billing-fraud identity-compromise cloud gcp llmjacking vertex-ai
5t 17i updated
critical advisory

Weaponization of Google Vertex AI Agents

Researchers demonstrated that AI agents built on Google's Vertex AI can be compromised to exfiltrate data, create backdoors, and compromise infrastructure by abusing excessive permissions of the Per-Project, Per-Product Service Agent (P4SA).

cloud ai vertex-ai privilege-escalation
2r 8t