Brief
Darwinium and Raonsecure unveil AI‑focused identity security tools
New capabilities aim to monitor AI agents’ actions and link identity to tool usage across platforms.
By Felo News Desk · Published
Darwinium, a California‑based identity‑management firm, announced two new capabilities designed to extend authentication and fraud‑prevention controls to autonomous AI agents, while Seoul‑based Raonsecure introduced a suite of five AI security products aimed at managing agent identities and threats.
What happened
On 9 October 2026, Darwinium released a statement describing its "Intent Intelligence" platform upgrade. The upgrade adds Journey Transition Probability, which flags atypical changes in the steps leading to a login, account action or payment, regardless of whether a human, bot or AI agent is involved. It also adds MCP Protection, which links an agent’s tool calls to the journey that led to them, allowing verification of credentials and additional checks for high‑risk transactions.
In a separate interview, Raonsecure CEO Lee Sun‑hyung detailed five new tools: ONEAccess for managing AI agent identities and privileges, ONEShield for real‑time AI‑targeted threat blocking, ONETag for automated document classification, ONESOAR for natural‑language‑driven security‑operations automation, and ONEHacker for AI penetration testing. Lee said the company plans to integrate these tools into a single security framework covering all stages of AI agent activity.
What the reports add
Darwinium’s release emphasizes that its technology assesses the order and timing of actions against a customer’s typical behavior, the business’s traffic patterns and the type of journey underway, flagging activity that may appear suspicious only in the context of the full path. The company notes that even an authenticated human can be coached into fraud, and that an authorized AI agent can deviate from its original task.
Raonsecure’s interview highlights that the five tools are intended to create a comprehensive defense against AI‑driven threats, from identity management (ONEAccess) to penetration testing (ONEHacker), and that the suite will be linked to form a unified security framework.
What was said
"Journey Transition Probability flags unusual changes in the steps leading to a login, account action or payment, whether a human, bot or AI agent is involved," Darwinium said in its release.
Michael Rodriguez, chief operating officer at Darwinium, warned, "An authenticated (human) customer can still be coached into sending money to a scammer. An authorized AI agent can start out doing exactly what a customer asked, then take an unexpected turn." (biometricupdate.com)
Lee Sun‑hyung told Chosun Biz that Raonsecure "plans to link these products to build a security framework that spans all stages." (biometricupdate.com)
Key facts
- Darwinium introduced Journey Transition Probability and MCP Protection to monitor AI agent behavior. (biometricupdate.com)
- Raonsecure announced five AI security tools: ONEAccess, ONEShield, ONETag, ONESOAR and ONEHacker. (biometricupdate.com)
- Michael Rodriguez said AI agents can deviate from their original tasks after gaining access. (biometricupdate.com)
Sources
- [1] biometricupdate.com — originally reported as “AI agents expose the gap between identity and authority”
