# Exploited LiteLLM, Artifactory and Switchvox Flaws Put AI and DevOps Infrastructure in Attackers’ Sights

*Thursday, September 3, 2026 at 6:13 AM UTC — Hamer Intelligence Services Desk*

**Published**: 2026-09-03T06:13:48.658Z (1h ago)
**Category**: cyber | **Region**: Global
**Importance**: 8/10
**Sources**: OSINT
**Permalink**: https://hamerintel.com/data/articles/16669.md
**Source**: https://hamerintel.com/summaries

---

**Deck**: Attackers are exploiting vulnerabilities in LiteLLM, Artifactory, and Switchvox to deploy crypto miners, open reverse shells, mint admin tokens, and harvest API keys, prompting U.S. authorities to add seven bugs linked to these attacks to a must-fix list.

A set of software flaws in tools that sit behind the scenes of many organizations’ IT and AI systems has moved from theory to active exploitation, forcing U.S. cyber authorities to elevate them as urgent risks.

Recent reporting describes attackers exploiting vulnerabilities in LiteLLM, Artifactory, and Switchvox. The intrusions are being used to deploy cryptocurrency miners, open reverse shells for remote control of systems, mint administrator tokens, and steal API keys.

In response, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added seven exploited vulnerabilities tied to these products to its Known Exploited Vulnerabilities (KEV) catalog. That list directs U.S. civilian agencies, and strongly influences many private organizations, to prioritize patching or mitigation.

The three products occupy important positions inside modern networks. LiteLLM acts as a layer between applications and large language models, handling API traffic that can include sensitive data and credentials. Artifactory manages software artifacts used across development and deployment pipelines. Switchvox underpins enterprise communications.

When these systems are compromised, attackers can gain leverage over a wide range of operations. Access to API keys can open paths into cloud services or other applications. Control of artifact repositories can allow tampering with software components before they are released. A foothold in a communications platform can support eavesdropping or lateral movement across internal networks.

The inclusion of the related vulnerabilities in CISA’s KEV catalog signals that exploitation is already occurring at scale. Organizations using LiteLLM, Artifactory, or Switchvox now face pressure to identify exposed instances, apply patches where available, and review logs for signs of compromise.

Key developments to watch include whether these flaws begin appearing more broadly in ransomware or state‑linked campaigns, and whether vendors or regulators respond with additional guidance on securing AI middleware, software supply‑chain tools, and communications platforms.