Research explained

Cloud AI inventories need to include model access and connected data

Qualys describes its inventory approach in a product article, rather than disclosing a new cloud-model vulnerability.

Published 29 Oct 2026 · Material update 28 Jul 2025 — last time the underlying evidence changed · Evidence: medium

What happened

A July 2025 Qualys article presents TotalAI’s approach to identifying cloud-hosted models and AI workloads. It discusses visibility across services such as Bedrock, Azure AI Foundry and Vertex AI.

These services are examples in a vendor product discussion, not confirmed affected products in a shared prompt-injection incident. The source should be read with that commercial context.

An effective inventory records who owns each deployment, which identities can invoke it, what information it can access and which configuration changes need review. Evaluate inventory tools against those requirements rather than assuming model discovery alone establishes security.

Read the original source.

Next step

Include identities, data access and ownership in your AI service inventory.

What remains unknown

Open question
The sources reviewed do not establish whether your deployment was exploited.

Source references