What matters in AI.

Subscribe

Honeypot sensors record 1,423 tries to use a Langflow vulnerability

A user who has a login can execute commands on the server in versions before 1.9.0.

Previdian found that attackers use a vulnerability in Langflow, an open-source platform for AI programs. Its honeypot sensors recorded 1,423 tries from 27 IP addresses since 19 September 2026. The vulnerability, CVE-2026-105740, has a CVSS score of 9.9. On 11 October 2026, the sensors recorded 333 tries. A user who has a login can execute commands on the server.

Sources

  1. Langflow's Critical Authenticated RCE Flaw Targeted in 1,423 Honeypot Exploit Attemptsthehackernews.com
AI MATTER · NEWS · AI MATTER · NEWS ·15 OCT2026

Posted

More in Security

All Security news