The corporate org chart is about to get a lot more crowded. Within a few years, your company may employ more AI agents than human beings. That shift is coming, and it is breaking the security systems that hold the modern enterprise together.

NewCore, a cybersecurity startup emerging from stealth today, is betting that the future of work requires a new way to manage these digital employees. The company announced a $66 million seed round led by Cyberstarts, with participation from Index Ventures and Evolution Equity Partners. The funding values the firm at $300 million. It is a massive bet on a simple premise: AI agents are not just software tools. They are workers.

Existing identity platforms like Okta or Microsoft Entra were built for people. They manage passwords, permissions, and access logs for humans. When you add thousands of autonomous AI agents to that mix, those legacy systems buckle under the complexity. Zohar Alon, NewCore’s CEO and a veteran of the cloud-security space, argues that treating an AI agent like a standard service account is a recipe for disaster.

“We know for sure that the scale and the complexity that those things are going to add to 15- or 20-year-old identity platforms are going to break them,” Alon said.

A Workforce of Code

Companies are already treating AI as staff. McKinsey reports that 25,000 AI agents are currently working alongside its human employees. Goldman Sachs has tested coding agents like Devin to handle real development tasks. These aren't just chatbots. They are agents that execute code, access databases, and make decisions.

NewCore’s platform treats these agents as first-class citizens. It provides them with unique identities, specific permissions, and a lifecycle that can be audited or revoked. Instead of relying on manually distributed credentials—which are often insecure and hard to track—NewCore uses a "split-key" architecture. This divides critical credentials between the customer and the platform, removing a single point of failure.

For developers, the platform integrates directly with tools like Anthropic’s Claude Code and OpenAI’s Codex. It allows these assistants to access enterprise systems securely. If an agent goes rogue or needs to be decommissioned, a manager can pull the plug via a mobile app. It is human oversight for a machine-driven workforce.

Why the Timing Matters

Identity is the new perimeter. If an AI agent has the keys to your production environment, it effectively has the keys to the company. Current vendors are trying to patch this by bolting agent support onto existing human-centric platforms. Alon believes that is a mistake. He argues that you cannot simply extend a system designed for humans to accommodate a workforce that operates at machine speed.

NewCore has already grown to 50 employees across the U.S. and Israel. While they currently have fewer than 10 customers, they are working with more than 10 design partners to refine the platform. They expect to begin charging for the service this summer.

Key Takeaways

  • AI as Employees: Companies are moving from using AI as a tool to deploying autonomous agents that act as staff members.
  • Identity Crisis: Legacy identity providers are struggling to manage the scale and speed of thousands of non-human agents.
  • New Guardrails: NewCore’s platform treats AI agents as distinct identities with their own permissions and human-in-the-loop oversight.

The Next Frontier

The transition is inevitable. As AI agents become more autonomous, the risk of credential theft or unauthorized access grows exponentially. The question is no longer whether AI will join the workforce. It is whether companies can build the guardrails fast enough to keep them under control. We will find out when NewCore’s first paying customers go live this summer.