CVE-2026-119990: Azure OpenAI Elevation of Privilege Vulnerability

Microsoft MSRC

This story is part of the morning brief Fireworks AI breached again through its cloud provider, and Anthropic won't fix a planted-file flaw in Claude's Office skills. The reporting lives at the original source.

Read at Microsoft MSRC

A second Fireworks incident exposes billing data and some customer model weights, a Claude skills flaw runs planted code on Linux, and a jailbroken agent spends two weeks inside a telecom.