NewCore raises $66M to give AI agents corporate IDs

Startup NewCore has emerged from stealth and raised $66M to tackle a problem few have put on the agenda yet: managing the identity of AI agents in the corporate environment. The founders argue that this, rather than defending against hackers or monitoring employees, will become the main enterprise security challenge in the coming years.

The AI agent as a new employee

AI agents at large companies are no longer pilot projects. They rewrite code, handle customer requests, execute transactions and manage supply chains. A single agent can act across dozens of corporate systems at once: reading databases, initiating payments, sending emails.

This is where the problem NewCore calls critical arises. Traditional identity and access management (IAM) systems were built for people: every employee has a unique name, a role, a history of actions and accountability. An AI agent has none of that in the usual sense. It can switch context, work under several names and act without a clear link to a specific human operator.

An ID for an algorithm

NewCore is building infrastructure that assigns every AI agent a verified digital identifier, a functional equivalent of a corporate badge for a software entity. The idea is that any company should be able to know exactly which agent has access to which resources, what it is doing right now and who is responsible for its actions.

The platform handles several tasks at once:

• Unique identification of every agent within an organization
• Granular control of access rights to systems and data
• A full audit trail: what the agent did, in what context and on whose instructions
• Integration with existing SOC, SIEM and IAM tools
• Detection of anomalous requests and potential data leaks through agents

A $66M round is a serious signal: investors consider the problem real and the potential market large enough to bet on now.

Why the moment has come now

Several factors have converged. OpenAI is pushing Operator, Anthropic Claude Agents, Google Gemini Workspace Actions, and Microsoft Copilot agents in its enterprise suites. Fortune 500 companies are increasingly deploying agents in critical business processes: finance, compliance, HR and legal.

"The next enterprise security challenge is managing AI agents, not people," is

NewCore's position.

Most companies have not yet built policies for this scenario. An agent can get temporary access to sensitive data and use it, and no internal audit will record this in connection with a specific business task or a responsible person. Regulators in the EU and the US are already eyeing this gap, and new rules are inevitable.

What this means

NewCore is betting on a new category: Agent Identity Management as a standalone discipline of corporate security. If the company's forecast proves right, the next three years will determine who becomes the Okta or CrowdStrike of the agentic era. For CISOs, this is a signal to start taking inventory now: how many AI agents run in your infrastructure, and who is responsible for them.