December 30, 2025
With the final release version in 2025, Saviynt is consistently expanding the Identity Cloud into an AI-native security platform. The company is thus addressing a central challenge of digital transformation: the parallel protection of AI-supported systems while simultaneously using artificial intelligence to strengthen its own security infrastructure.
Identity security in the changing AI era
Table of Contents
The past year marks a turning point in corporate IT. Artificial intelligence has evolved from experimental pilot projects to productive core systems. Companies are increasingly using autonomous agents and processing data on an unprecedented scale.
This development brings with it a new category of security risks. Non-human identities – service accounts, API tokens, automation scripts – already outnumber human identities by a ratio of 82:1. AI agents themselves are becoming independent actors that require governance structures. Conventional security concepts reach their limits when it comes to controlling these dynamic, autonomous systems.
Dual strategy: Security for and through AI
Saviynt is taking a two-pronged approach with the current version. On the one hand, the platform is intended to secure AI systems and, on the other hand, to use AI technology to improve identity security.
Identity Security Posture Management for all identity types
The expanded ISPM functions now cover the entire spectrum of identities – from classic employee accounts to external access to non-human identities and AI agents.
The new ISPM module for non-human identities gives organizations complete transparency across all service accounts in the company for the first time. The solution visualizes relationships between service accounts, their resource access, credentials and associated permissions. A native ServiceNow integration enables direct remediation workflows.
For AI agents, Saviynt offers a specialized ISPM framework. These autonomous units not only access systems, but also make independent decisions and interact with complex infrastructures. The framework identifies all AI components including agents, MCP servers and tools. Companies can track complete access paths and document changes via timeline views. This means that every autonomous action remains controllable and auditable.
Agent-based automation for application onboarding
Saviynt addresses a central problem in the modern IT landscape with the new Integration and Onboarding Agent. Almost half of all companies lose track of unconnected applications – a significant security risk due to a lack of governance.
The Integration Agent is based on Computer Using Agent technology and is therefore fundamentally different from rigid RPA bots or simple NLP tools.
The solution interacts with applications like a human administrator, recognizes user interfaces and adapts to changing interfaces. Integration projects that previously took weeks or months can now be implemented within a few hours – without sacrificing stability or governance quality.
Any web-based application can be integrated, regardless of API availability or traditional integration methods.
Platform optimizations for productive use
In addition to the AI-specific innovations, Saviynt has further developed the platform in several areas.
Redesigned user interface
Continuous UI/UX updates make working with the platform easier. The optimizations are aimed at clearer workflows and more intuitive controls. Security teams can navigate through processes, interpret data and initiate time-critical measures more efficiently.
Advanced ecosystem integration
Saviynt increases interoperability with other security solutions. New interfaces to endpoint security platforms, network security systems, data security tools and cloud security solutions enable bidirectional data exchange.
The platform aggregates information from the entire security stack and enriches external systems with identity context. For example, endpoint security decisions are supplemented with real-time identity risks. Data security integrations bring data protection status into identity governance workflows. By correlating identity data within the Saviynt platform and enriching external security tools, comprehensive visibility is created.
Security teams can see connections between identity events and broader security incidents and move from reactive incident response to proactive risk management.
Outlook for 2026
Saviynt is planning further expansion stages for the coming year. Announced are expanded posture management functions for MFA and SSO as well as deeper governance mechanisms for AI agents and MCP servers. In addition, advanced risk analysis, identity verification functions and specialized agent-based AI solutions will be provided for defined user groups.
Integration into the security ecosystem continues to deepen to keep pace with the speed of modern enterprise environments. The provider is positioning itself as a platform for managing all types of identities – human and non-human.
Saviynt ISPM now supports posture management for your non-human and AI agent identities. Those: Savage
More reading material:
