NewCore, an agentic identity security company, has received recognition in the Enterprise Security Tech Cyber Top Innovations Awards for its approach to governing and securing enterprise AI agents. The recognition highlights NewCore Agent Guardian, a security solution designed to give autonomous and semi-autonomous AI agents dedicated identities, controlled permissions and traceable accountability. As businesses deploy AI agents capable of accessing sensitive applications, interacting with enterprise systems and executing tasks independently, NewCore is positioning identity governance as a critical layer of enterprise AI security.
NewCore Brings Identity Governance to Enterprise AI Agents
The rapid adoption of autonomous AI agents is creating a new challenge for enterprise cybersecurity teams: controlling digital workers that can access applications, retrieve sensitive information and execute actions across multiple systems.
Traditional identity and access management (IAM) platforms were largely designed around human employees, service accounts and applications with established permission structures. AI agents introduce a more dynamic security environment. They can initiate workflows, invoke tools and perform operations at machine speed, sometimes without direct human supervision at every step.
NewCore is addressing this challenge with Agent Guardian, an identity governance and security solution designed specifically for agentic AI environments. The company recently received recognition for the technology in the Enterprise Security Tech Cyber Top Innovations Awards.
The platform’s central proposition is to treat every AI agent as a distinct enterprise identity, with a defined owner, permissions, lifecycle and accountability framework. Rather than simply authenticating agents when they connect to systems, NewCore focuses on governing what they are authorized to do throughout their execution.
Agent Guardian Applies Least-Privilege Access to AI Agents
One of Agent Guardian’s core capabilities is assigning each AI agent an individual identity linked to a responsible human owner.
The platform uses a live identity graph to discover agents, accounts, applications and relationships between their access permissions. This is intended to give security teams visibility into which agents operate within enterprise environments and what resources they can access.
NewCore also applies just-in-time credential management and least-privilege access controls. Instead of giving agents broad, persistent credentials, the system provides narrowly scoped, short-lived credentials for authorized tasks.
This approach addresses a significant security concern associated with autonomous systems: an agent that receives excessive permissions can potentially use those privileges beyond its intended purpose, particularly when interacting with multiple tools or applications.
The platform also evaluates policies at runtime. Every agent tool call can be assessed against established rules, allowing permitted activities to proceed while blocking prohibited operations.
NewCore says administrators can create policies using natural language, with the platform identifying potential policy conflicts before deployment. This could make governance easier for security teams managing large numbers of agents and increasingly complex authorization requirements.
Human Approval Adds Oversight to High-Risk Agent Actions
Another important component is NewCore’s human-in-the-loop approval mechanism.
When an AI agent attempts a sensitive operation, Agent Guardian can send an approval request to its designated owner through a mobile device. The owner can approve or deny the action using biometric verification.
This creates a distinction between routine automated operations and activities that require human judgment. Routine actions can proceed under predefined policies, while higher-risk operations can be escalated for explicit approval.
The platform also maintains audit trails connecting each agent action to its responsible owner, the policy applied and the approval decision. Security teams can revoke access immediately when an agent exceeds its authorized boundaries or no longer requires access.
These controls address several challenges associated with agentic AI security:
- Identity visibility: Discovering agents and understanding their relationships with enterprise applications.
- Credential security: Eliminating persistent secrets and limiting access to specific tasks.
- Runtime authorization: Evaluating permissions during execution rather than relying exclusively on initial authentication.
- Human accountability: Establishing ownership and approval requirements for sensitive operations.
- Auditability: Maintaining records that help security teams investigate agent activity and policy violations.
NewCore delivers these capabilities through a unified identity platform covering authentication, authorization, security and governance for both human and non-human identities. The company says Agent Guardian can operate alongside an organization’s existing identity platform, potentially reducing the need to replace established identity infrastructure.
AI Agent Expansion Is Increasing Identity Security Risks
The timing of the recognition reflects a growing enterprise security challenge: AI agents are becoming operational participants rather than simply tools that generate recommendations or content.
Gartner’s April 2026 research predicts that an average Fortune 500 enterprise will have more than 150,000 AI agents in use by 2028, compared with fewer than 15 in 2025. The research also found that only 13% of organizations believe they have appropriate AI agent governance in place. Gartner
This projected expansion creates pressure on identity and access management systems. Every additional agent can introduce another identity, credential, permission set and potential route to sensitive enterprise resources.
The broader cybersecurity market is responding with approaches that extend traditional identity governance into machine identities, workload identities and autonomous AI systems.
Established technology providers, including Microsoft, Google and Amazon Web Services, are developing security and governance capabilities across their enterprise AI ecosystems. Identity security specialists are also addressing the need to control non-human access and privileged operations.
NewCore’s approach focuses specifically on bringing human and AI identities under one governance architecture, with authorization enforced at the individual action level.
The Cyber Top Innovations recognition provides industry visibility for that approach, although an award does not independently establish security effectiveness or demonstrate protection against every agent-related threat.
For enterprises evaluating agentic AI security platforms, the more important measures will be policy enforcement reliability, credential protection, integration with existing identity systems and the ability to trace actions back to accountable owners.
As autonomous systems gain access to more enterprise applications and data, identity governance is becoming a foundational requirement for scaling AI agents responsibly. NewCore is positioning Agent Guardian around that requirement, with the goal of allowing organizations to automate more work without surrendering control over who—or what—can act on their behalf.
Market Landscape
Enterprise AI security is evolving as organizations move from conversational AI tools toward autonomous agents capable of executing business operations.
Five developments are shaping this market:
- Agentic identity management: Enterprises need dedicated identities, ownership and lifecycle controls for autonomous AI agents.
- Non-human identity security: Service accounts, machine identities and AI agents are expanding the traditional identity attack surface.
- Zero-trust authorization: Just-in-time credentials, least-privilege permissions and continuous authorization are becoming important security requirements.
- Human-in-the-loop governance: Sensitive AI actions require approval mechanisms that preserve human accountability without slowing routine automation.
- Runtime AI security: Security platforms are increasingly evaluating agent actions during execution rather than relying exclusively on initial authentication.
Gartner’s forecast of more than 150,000 AI agents in an average Fortune 500 enterprise by 2028 illustrates the potential scale of the governance challenge.
NewCore is entering this market with a unified identity architecture designed to govern humans and AI agents together. Its competitive positioning will depend on policy enforcement, credential security, integration capabilities and visibility across increasingly complex agentic environments.
Top Insights
- NewCore Agent Guardian received recognition in Enterprise Security Tech’s 2026 Cyber Top Innovations Awards for AI agent identity governance.
- The platform assigns individual identities, human ownership and lifecycle controls to enterprise AI agents.
- Just-in-time credentials and runtime authorization restrict agents to narrowly scoped permissions for specific tasks.
- Biometric human approval provides additional oversight for sensitive AI agent operations requiring explicit authorization.
- Gartner predicts average Fortune 500 enterprises will operate more than 150,000 AI agents by 2028, increasing identity governance requirements.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
