PNP Secure Presents AI Agent Control Strategy at FISCON 2026
IT DAILY ·
✦ AI Summary
PNP Secure announced on the 13th that it introduced its AI agent control strategy at the Financial Information Security Conference (FISCON) 2026.
It said the expansion of work automation has increased the need for access management for AI agents and automation tools.
It explained a zero-trust implementation direction that combines facial authentication and physical input verification and applies it to PC access, business system use, DB and server access, and command execution.
BizCrushWant to follow Korean talks and live streams in real time?Try CapCrush on the web
PNP Secure announced on the 13th that it introduced its AI agent control strategy at the Financial Information Security Conference (FISCON) 2026. The event was held on the 12th at the Conrad Seoul Hotel in Yeouido, Seoul.
PNP Secure said that the expansion of work automation has increased the need for access management for AI agents and automation tools. In particular, it identified remote control using legitimate accounts or compromised sessions as a control target, and explained that account validity verification is a necessary condition, followed by an additional requirement to confirm who is actually using the authority.
Accordingly, PNP Secure presented a strategy that combines facial authentication and physical input verification. As proposed principles, it introduced a method to confirm through facial information whether an authorized user is physically present in front of the terminal, and to verify whether physical keyboard and mouse input accompanies access and work.
PNP Secure explained a zero-trust implementation approach that applies these principles through product integration by stage across PC access, business system use, database(DB) and server access, and command execution sections. The on-site photo shows the PNP Secure booth at FISCON 2026 on the 12th.
The booth demonstration focused on access control based on FaceLocker, and was centered on a scenario in which automated attacks using generative AI infiltrate an internal server, then attempt subsequent access and work on other servers after the breach. When accessing files, folders, and servers, the system was configured to verify physical input, user presence, facial information, and access privileges, and it demonstrated a policy-based restriction process when the verification conditions were not met. The goal was to control unauthorized access paths that extend to assets after an internal intrusion.
At the same time, as a means of supporting convenience at work, the company also presented a method of linking facial authentication with existing single sign-on(SSO) systems and individual business systems. Through this, it proposed a direction that strengthens user verification while reducing the burden of repeated authentication. It also introduced a screen information leak prevention function, and the targets of response are smartphone photography and third-party access.
A PNP Secure official said that as AI agents and automation tools spread further into work environments, control that verifies both account privileges and the actual operator becomes more important. The official added that the company plans to support the phased implementation of enterprise-specific zero-trust control systems centered on facial authentication and physical input verification.
Source: IT DAILY · Kim Ho-jun
Original: https://www.itdaily.kr/news/articleView.html?idxno=242919
References
This article was produced with the help of an automated content generation algorithm.
Source: IT DAILY
View originalThis article was summarized and organized by BizCrush based on the original article from IT DAILY. For exact quotations and full details, please refer to the original article.
Keep reading
BizCrush