{ads}

›

ServiceNow AI Agents Can Be Tricked Into Acting Against Each Other via Second-Order Prompts

Written By Wednesday, 19 November 2025 Add Comment



November 19, 2025 at 03:29PM

Malicious actors can exploit default configurations in ServiceNow's Now Assist generative artificial intelligence (AI) platform and leverage its agentic capabilities to conduct prompt injection attacks. The second-order prompt injection, according to AppOmni, makes use of Now Assist's agent-to-agent discovery to execute unauthorized actions, enabling attackers to copy and exfiltrate sensitive

from The Hacker News https://ift.tt/3dwgQCY

0 Response to "ServiceNow AI Agents Can Be Tricked Into Acting Against Each Other via Second-Order Prompts"

Post a Comment

Subscribe to: Post Comments (Atom)

Article Top Ads

Central Ads Article 1

Middle Ads Article 2

Article Bottom Ads