Skip to content
Threat Feed
Subscribe

Vendor

DeepSeek

3 briefs RSS
medium advisory

Emerging Trends in AI-Integrated Malware and Tracking via CAIRN

Cisco Talos has documented the emergence of malware leveraging LLM services and local inference, introducing the CAIRN framework to track these threats via metadata-based cognitive artifacts.

OpenAI API +3 ai-threats malware detection metadata-analysis infrastructure-tracking
3t 4i
critical advisory

Authentication Bypass in DeepSeek Harness via Host Header Spoofing

DeepSeek Harness versions prior to 0.1.2-alpha.1 contain an authentication bypass vulnerability allowing unauthorized remote control of the agent via a spoofed HTTP Host header.

Harness
2t 1c
high advisory

DeepSeek TUI SSRF Vulnerability via HTTP Redirect Bypass (CVE-2026-45310)

DeepSeek TUI is vulnerable to a Server-Side Request Forgery (SSRF) attack (CVE-2026-45310) because the `fetch_url` tool validates the initial URL against a restricted-IP blocklist but fails to re-validate redirect targets, allowing attackers to exfiltrate sensitive information from cloud-hosted instances by using a redirect to a restricted IP address.

deepseek-tui +1 ssrf prompt-injection cloud-metadata
2r 1t 3i