Feed›Malware›JadePuffer agentic attacks now target AI model data with ran...
MalwareBleeping Computer
9.8 — CRITICAL

JadePuffer agentic attacks now target AI model data with ransomware

📅 20 July 2026 at 21:08 UTC📰 Bleeping ComputerView original source ↗
JadePuffer agentic attacks now target AI model data with ransomware

The JadePuffer autonomous AI agent has upgraded with custom malware called EncForge that focuses on encrypting AI assets, such as training datasets, vector databases, and model checkpoints. [...]

🤖 AI BriefingAuto-generated threat analysis
🔍Threat Overview

JadePuffer, an autonomous AI agent threat actor, has upgraded its tactics with custom malware called EncForge that targets AI model data with ransomware, causing significant disruption and potential financial losses for affected organizations.

⚙️Technical Details
Affected Systems
Langflow Langflow
Attack Vectors
NETWORK
💥Impact Assessment
Severity: CRITICAL
Who Is at Risk
Organizations with AI model data, particularly those using Langflow version prior to 1.3.0
🛡️Recommended Actions
1Apply available security updates, most notably Langflow version 1.3.0 or later
2Restrict Docker socket access and run Langflow containers as non-root
3Apply filesystem-level access controls to model weight directories
📦Affected Products
Langflow LangflowLangflow
🔐NVD Verified DataVERIFIED
CVE-2025-3248 ↗CVSS 9.8 — CRITICAL
Attack Vector
NETWORK
Complexity
LOW
Vector String
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Weaknesses
CWE-94CWE-306
Affected Products (CPE)
Langflow Langflow

Read the full article

This is a curated summary. The complete article is available at Bleeping Computer.

Read on Bleeping Computer ↗
← Back to feed