Attackers Target AI Development Platform Langflow
F5, Tuesday, October 6th, 2026
F5 Labs saw 405 attempts to exploit Langflow RCE CVE-2026-0768, its first AI development platform RCE in mass scanning.
F5 Labs' October Sensor Intelligence Series reports that its honeynet recorded 405 requests from 55 source IPs targeting CVE-2026-0768, an unauthenticated RCE in the Langflow AI application-building platform.
It is the first AI/ML development platform RCE F5 Labs has seen in opportunistic internet-wide scanning, with some payloads testing code execution and others seeking environment data holding cloud credentials, model API keys and database secrets.
September also brought renewed exploitation of React Server Components and Microsoft Exchange vulnerabilities.