RaonSecure unveils security solutions for agentic AI

RaonSecure Co. has unveiled five security solutions designed to protect agentic artificial intelligence (AI) systems, covering identity management, data protection, and threat response as AI increasingly performs tasks autonomously.

RaonSecure introduced the solutions at its 2026 Security IQ Up (SQUP) event in Seoul on Wednesday, under the theme “Make Agentic AI Fun and Secure.”

RaonSecure Chief Executive Officer Lee Soon-hyung said AI security must extend beyond external safeguards as AI systems gain the ability to make decisions and access external systems.

“If you cannot identify it, you cannot control it,” Lee said. “AI security must address five areas: identity, behavior, data, defense and attacks.”

RaonSecure proposed using Web3 technology to give AI and robots unique digital identities and manage their permissions. The system can track whose instructions an AI is following, what authority has been delegated, and the scope of actions it is allowed to take.

The five solutions are OneAccess, which manages identities and permissions for AI agents and robots; OneHacker, which uses AI to identify potential attack paths; OneTag, which automatically classifies documents by security level; OneSOAR, which automates account management and threat response through natural-language commands; and OneShield, which blocks threats such as malicious prompts in real time.

Companies can adopt the solutions individually depending on their AI transformation, or AX, stage and security needs, RaonSecure said.

Kim Tae-jin, RaonSecure’s chief technology officer, said the company developed the solutions as an integrated security framework rather than as standalone products.

“The key is that these technologies work together as a single security framework covering the entire AI environment,” Kim said.

The event also featured a demonstration of assigning a digital identity to an AI system. RaonSecure’s humanoid robot introduced as the company’s first AI employee completed an authentication process to receive a digital employee ID before verifying visitors’ mobile IDs. The demonstration showed how companies could manage AI systems and robots similarly to human employees by assigning identities and permissions and limiting their actions to authorized tasks.

RaonSecure expects AI security and trust frameworks to become increasingly important as companies expand AI adoption, with competition shifting from AI development capabilities to safe and reliable use.

By Park Seong-bae and Chang Iou-chung
[ⓒ Pulse by Maeil Business News Korea & mk.co.kr, All rights reserved]