AI identity startup NewCore raises $66M

NewCore has emerged from stealth mode with $66 million in funding to tackle a challenge most enterprises haven’t fully articulated yet: determining exactly who or what is accessing their systems. The identity-security startup, headquartered in both Tel Aviv and San Francisco, is developing a platform that manages both human employees and autonomous AI agents within a single framework. The company is already valued at $300 million.

The financing actually consists of two rounds combined. A $16 million pre-seed was led by Index Ventures and Cyberstarts, followed by an expanded seed round led by Evolution Equity Partners that pushed the total to $66 million, all at that $300 million valuation.

The team’s background forms a major part of the appeal. CEO Zohar Alon previously founded Dome9, a cloud-security firm later acquired by Check Point. His co-founders include Amihai Neiderman, a former Unit 8200 research leader who also founded Nym Health, and Erez Yarkoni, who served as CIO for both T-Mobile US and Telstra. Prominent Israeli security founders like Wiz’s Assaf Rappaport and Cyera’s Yotam Segev have also invested as angel backers.

Here is what NewCore is actually building. The concept starts from a trend already in motion. Companies are increasingly deploying AI agents that operate independently, querying databases, moving funds, filing tickets. Each of those agents requires credentials, permissions, and an audit trail, just like a human employee. Currently, this creates chaos: agents either borrow human logins or run on static keys that nobody monitors.

NewCore aims to become the central system of record for all identities. It issues identities to agents, defines what they can access, and logs their activity, all within the same framework a company uses to manage its staff. Rather than treating the agent problem as an add-on, the company positions it as a reason to rebuild workforce identity from the ground up. That is why it claims to be going “head-to-head with Microsoft and Okta,” the dominant players in that space today.

The company has even shipped an “Agentic Skill” feature that allows coding agents like Anthropic’s Claude Code, OpenAI’s Codex, and Cursor to log into corporate systems as managed identities instead of using borrowed credentials.

Why are investors backing this so early? The product is not vaporware. NewCore is already live with fewer than 10 paying customers, more than 10 design partners, and plans to start charging this summer. The $300 million valuation is essentially a bet on timing. The number of non-human identities inside organizations is exploding, and security teams increasingly recognize that an AI agent with the wrong permissions represents a breach waiting to happen.

That fear, more than any specific feature, is what NewCore is really selling.

The challenge is that Microsoft and Okta are not passive observers, and the “identity for agents” category is quickly becoming crowded rather than open terrain. For now, NewCore’s advantage lies in its investors and the founders betting on it.

The real test is whether a security-first identity layer for agents will become its own standalone product, or whether the incumbents will absorb it as a feature before any startup can claim the space.