6 Key Security Risks in LLMs: A Platform Engineer’s Guide

More organizations than ever are deploying AI models, whether through managed cloud services or self-hosted solutions. According to The State of AI in the Cloud 2025 report from cloud security provider Wiz, 85% of organizations now use AI services, up from 70% in 2024. With this growth comes more integrations, tools and security risks.

Platform engineers are on the front lines of securing these AI systems, especially as large language models (LLMs) enter production. While cloud-hosted LLMs pose significant security challenges, self-hosted…