Key Takeaways
- Israeli cybersecurity veterans are launching startups to secure AI agents, bots, and other non‑human identities inside enterprises.
- Traditional identity‑and‑access‑management (IAM) tools were built for human users and cannot cope with the scale, continuity, and decision‑making autonomy of AI systems.
- Major investments and acquisitions—such as Cyera’s $1 billion purchase of Oasis Security—signal that securing non‑human identities is becoming a strategic priority for large security vendors.
- New entrants like NewCore, Oak, Linx Security, Venice, and several niche players are offering platforms that provide dynamic, context‑aware access controls across humans, machines, and AI agents.
- The emerging market addresses urgent questions: who created an AI agent, what data it can access, what actions it may perform, and how to revoke permissions when behavior deviates from policy.
The Shift from Human‑Centric to AI‑Centric Identity
For decades, enterprise security focused on answering a simple question: who is accessing a system? Identity‑and‑access‑management (IAM) solutions were built around authenticating employees, assigning static roles, and enforcing least‑privilege policies. The rise of autonomous AI agents, software bots, and machine‑to‑machine interactions has overturned that model. An AI system may need to touch dozens of applications, analyze financial data, modify code, or interact with customers—often continuously and at massive scale. Consequently, security teams must now ask not only whether an entity is authorized to enter a system, but also whether it should be allowed to perform a specific action at a given moment. This nuanced, context‑driven evaluation is beyond the capability of legacy IAM platforms, prompting Israeli entrepreneurs to rebuild identity security for the AI era.
NewCore’s Ambition to Challenge IAM Giants
In June 2026, NewCore emerged from stealth with a $66 million funding round, led by veterans of some of Israel’s most notable cybersecurity successes. Founders Zohar Alon (co‑founder of Dome9), Amihai Neiderman (founder of Nym), and Erez Yarkoni (former CIO of T‑Mobile and Telstra) set an ambitious goal: to compete directly with identity behemoths Microsoft and Okta by delivering an AI‑native IAM platform. NewCore argues that enterprises today operate a patchwork of identity tools designed for human users and static applications, leaving gaps when thousands of autonomous agents seek access. Its vision is a unified framework that can govern access for humans, machines, and AI agents alike, applying policies that adapt to the agent’s origin, intended behavior, and real‑time risk signals.
Oasis Security and the Focus on Non‑Human Identities
Before being acquired by Cyera for $1 billion, Oasis Security had already positioned itself as a specialist in non‑human identities—machine accounts, service accounts, applications, and AI agents. The startup raised $120 million in a Series B round in March 2026, highlighting the growing disparity between the number of human users and the proliferating software identities within large organizations. Oasis contended that traditional IAM tools assume direct human supervision and therefore fail when software can independently initiate actions, access sensitive data, and execute workflows. Cyera’s acquisition underscored that securing these non‑human identities is no longer a niche concern but a core component of enterprise security strategy, prompting larger vendors to invest heavily in the space.
Oak’s AI‑Native Identity Operating System
Another notable entrant, Oak, closed a $60 million round in July 2026. Founded by former Torq CEO Shai Morag and Tal Marom, Oak markets itself as an AI‑native identity operating system capable of managing employees, machines, and AI agents from a single console. The company points out that existing IAM infrastructure has become a tangled assortment of disconnected tools—each handling a slice of authentication, authorization, or governance—incapable of coping with the complexity introduced by autonomous systems. Oak’s platform seeks to answer critical governance questions: Who created the AI agent? What data may it access? Which decisions is it permitted to make? How can permissions be instantly revoked if the agent exhibits anomalous behavior? By providing dynamic, policy‑driven controls, Oak aims to replace static role‑based models with continuous trust evaluation.
Linx Security, Venice, and Niche Innovators
Linx Security, which raised $50 million in March 2026, likens its mission to what Wiz did for cloud security: delivering a comprehensive platform that surfaces and secures the new layer of risk introduced by AI agents and non‑human identities. Venice, founded by Rotem Lurie and backed by $25 million in February 2026, focuses on replacing static enterprise permissions with dynamic access controls that shift in real time based on contextual risk factors such as user behavior, data sensitivity, and threat intelligence. A cohort of narrower‑focus startups further illustrates the market’s diversification: Hush Security targets the AI workforce, Offroad builds an AI identity security team, Tenet protects autonomous agents accessing sensitive workflows, and Capsule Security concentrates on securing enterprise AI agents. Each addresses a slice of the broader challenge—provisioning, monitoring, governance, or remediation—yet together they signal the emergence of a distinct cybersecurity category centered on AI governance and identity.
Why the Market Is Accelerating Now
The surge in funding and strategic acquisitions reflects several converging trends. Enterprises are moving AI from isolated pilots to core business processes, increasing the volume and criticality of non‑human identities. Regulatory pressure is mounting to ensure accountability for automated decisions, driving demand for auditable access logs and enforceable policies. Simultaneously, the sophistication of threat actors has grown; compromised AI agents can exfiltrate data, manipulate financial models, or disrupt operations at scale, making traditional perimeter defenses insufficient. Israeli founders, drawing on deep expertise from prior cybersecurity victories, are capitalizing on this gap by building platforms that combine identity verification, behavioral analytics, and adaptive policy enforcement. Their collective effort suggests that the next frontier of enterprise security will not be keeping hackers out of the network, but ensuring that every entity—human or machine—inside the network acts only as authorized and intended.