Neo Emerges From Stealth With $100M to Control and Secure Enterprise Agentic AI
Neo Security exited stealth this week with $100M in total funding, the largest single data point yet on how much capital is chasing agentic-access governance. The raise breaks down as a $75M Series A co-led by Andreessen Horowitz and Bessemer Venture Partners — with Craft Ventures and Merlin Ventures participating — on top of a previously undisclosed $25M seed from 2025.
The founding bench is notable: CEO Nick Warner is the former president/COO of SentinelOne, CPO Shlomi Salem led detection engineering there, and Eran Shirazi is CTO. Angels include Wiz’s Assaf Rappaport.
The product pitch is a real-time control layer for agentic software: inventory the tools and agents running across an enterprise, assess whether they’re safely configured, and enforce policy before risky activity occurs. The hard problem here is that agents typically act under a valid user’s permissions — so the exposure isn’t a stolen credential, it’s a legitimately authorized agent doing something it shouldn’t with data it can already reach.
Why data-protection teams should care. This is not a classic DLP or DSPM engine — it bridges endpoint, network, and identity rather than scanning stores or watching egress channels. But the data-exposure angle is direct. Neo’s team seeded credibility with March research exposing an accounting-firm leak that surfaced schemas, API keys, and auth tokens — exactly the machine-readable, high-value data that autonomous agents brush up against. As agents get write access to CRMs, code repos, and file stores, the question of “what data can this agent touch, and under whose authority” becomes a data-security posture problem whether or not it lives in a DSPM console.
Context for the shortlist. Neo lands in the same lane as a string of recent agentic-security deals and acquisitions — Straiker, Archestra, Zscaler’s Symmetry buy, and Cisco’s Astrix pickup among them. For buyers, the practical takeaway is that agentic-access governance is consolidating into a distinct budget line adjacent to, but not yet inside, DLP and DSPM. Expect overlap and turf negotiation with existing data-security tooling as these platforms mature.