Prompt-injection flaw found in $4B agentic AI app Manus

A prompt-injection vulnerability has been identified in Manus, a $4B-valued agentic AI application, allowing attackers to manipulate the AI's behavior through crafted external data. The flaw underscores systemic security risks in AI systems that process untrusted input. Impact and exploitation status remain unclear, but the incident highlights a growing attack surface in enterprise AI deployments.