NewCore has emerged from stealth with $66 million in funding to solve a fast-growing enterprise problem: how to give AI agents identities, permissions, and governance controls when they begin acting like digital employees. The launch matters because companies are moving from chatbots that answer questions to agents that use tools, access systems, and make workflow decisions.

What Happened

According to TechCrunch, NewCore’s funding includes backing from Cyberstarts, Accel, and other investors. The company’s pitch is that AI agents need a dedicated identity and security layer because they are not ordinary software accounts and they are not human employees. They can operate across SaaS tools, internal databases, code repositories, ticketing systems, and cloud environments, often with permissions that were originally designed for people or service accounts.

That mismatch creates a new security surface. If an agent has too much access, a prompt injection, compromised workflow, or poorly scoped automation can cause real damage. If an agent has too little access, it cannot complete useful tasks. NewCore is trying to occupy the middle ground: assigning identities to agents, tracking what they do, enforcing policy, and giving security teams a way to manage non-human AI actors as they scale.

Why It Matters

Enterprise AI adoption increasingly depends on trust infrastructure. Companies can experiment with copilots relatively safely when the AI is only drafting text or summarizing documents. The risk profile changes when an AI agent can approve a refund, query customer records, update production systems, send emails, or run code. At that point, identity and access management become core AI architecture rather than back-office security plumbing.

The rise of agentic AI also changes accountability. With a human employee, organizations can tie actions to a person, a role, and a manager. With AI agents, the chain of responsibility can become murkier: the agent may be created by one team, powered by a third-party model, connected to multiple tools, and triggered by a user prompt or automated workflow. A purpose-built control plane can help companies answer basic but critical questions: which agent acted, under whose authority, with what permissions, and why?

The Enterprise Security Angle

NewCore’s timing reflects a broader shift in cybersecurity. Identity security vendors, cloud providers, and AI infrastructure startups are all trying to adapt to a world where software can initiate complex actions on behalf of users. Traditional service accounts are often long-lived, overprivileged, and poorly monitored. AI agents make that old pattern more dangerous because their behavior can be dynamic and context-driven.

The company’s challenge will be integration. Enterprises already use identity providers, privileged access systems, cloud security tools, data-loss prevention platforms, and logging pipelines. To become a default layer for AI agents, NewCore will need to fit into that stack without adding too much friction. If it succeeds, agent identity could become a standard procurement requirement for companies deploying AI at scale.

Key Takeaways

• NewCore launched with $66 million to build identity and governance tools for AI agents.
• The company is targeting enterprises that want agents to act across real business systems.
• Agent identity is becoming a central part of AI security and compliance planning.
• The category could grow as companies move from copilots to autonomous workflows.

Source: TechCrunch