KUALA LUMPUR, March 19 (Bernama) -- Menlo Security, the leader in human and agentic Browser Security, has unveiled the first Browser Security Platform purpose-built to secure the agentic enterprise, where autonomous artificial intelligence (AI) agents will outnumber human employees and the browser has become the operating system for both.
Menlo in a statement said the platform provides the first unified control plane that applies machine-speed governance and threat prevention to both human and AI actors, deployed globally on Menlo’s elastic cloud infrastructure.
“By moving protection directly into the browser session, we are enabling organisations to deploy AI agents that work at a scale and speed impossible for humans without opening the door to catastrophic prompt injection or data exfiltration.
“Without this protection, a single compromised AI agent can move laterally across enterprise systems, exfiltrate data, or execute fraudulent transactions at machine speed, with no human in the loop,” said Menlo Security Chief Executive Officer, Bill Robbins.
This launch follows a record fiscal year for Menlo, which surpassed US$140 million in annual recurring revenue with net retention exceeding 120 per cent, alongside a strategic partnership with Google to provide least-privileged remote access to desktop applications and data via the browser for both humans and AI agents. (US$1=RM3.93)
The Menlo Browser Security Platform provides Menlo AI Agent Security that enforces instruction-data separation to ensure agents never mistake malicious data for legitimate commands and are unable to exfiltrate sensitive data.
It also facilitates secure access to applications and data, sanitising inputs so agents can safely replicate human workflows. With least-privileged agent governance, the platform enforces granular controls to prevent autonomous agents from moving laterally or accessing unauthorised data.
By embedding security directly into the browser session, it creates “Architectural Immunity”, neutralising evasive threats before they reach AI agents or human endpoints and providing a unified trust layer for the enterprise.
-- BERNAMA
