Zenity’s $125M Series C: Securing the Coming Era of 1 Billion AI Agents

A Security Category Built Entirely Around Agent Behavior

Zenity announced a $125 million Series C funding round on August 4, 2026, led by Norwest with participation from new strategic backers Qumra Capital, SoftBank Vision Fund 2, Hitachi Ventures, and LG Technology Ventures, alongside returning investors Vertex Ventures, Third Point Ventures, DTCP, and Intel Capital, according to SecurityWeek’s coverage of the round. The new capital brings Zenity’s total funding to $180 million, per fintech.global’s report on the deal.

Zenity, founded in 2021, does not sell a general-purpose security product — its platform is built specifically to govern autonomous AI agents, the software systems that access enterprise data, invoke tools, and execute multi-step workflows without a human approving each action. SecurityWeek describes the core mechanism as real-time analysis of an agent’s intended actions, which the platform can then deterministically allow, modify, or block before execution — rather than reviewing logs after an agent has already acted. fintech.global reports the platform covers major agentic AI frameworks including Microsoft Copilot, ChatGPT Enterprise, Gemini, and Claude, plus custom-built enterprise agent systems, positioning it as a cross-vendor governance layer rather than a tool tied to one AI provider.

The company runs commercial operations from New York with more than 230 employees worldwide, according to fintech.global. Zenity Labs, its internal research arm, is dedicated to finding vulnerabilities in agentic AI platforms before attackers do — a research function that doubles as a pipeline for new detection capabilities in the commercial product.

The Growth Numbers Behind the Round

fintech.global reports that Zenity’s revenue has tripled in each of the last two years, with the company projecting the same trajectory for the current year. Its customer base already spans most of the Fortune 500 and Global 2000, according to the same report — a client concentration that signals large enterprises are already treating agent governance as a budgeted security line item rather than an experimental purchase.

Zenity co-founder and CEO Ben Kliger framed the round against the pace of enterprise AI adoption: in a statement reported by fintech.global, Kliger said “enterprise AI is skyrocketing” and that “AI experimentation is long over” as organizations deploy agents “at velocity and adoption rates never seen before in any tech wave.” Norwest partner Assaf Harel, quoted by SecurityWeek, described Zenity as having “an early mover advantage in building end-to-end AI agent security” with the largest footprint among large enterprise customers of any agent-security specialist.

That framing matches a broader pattern visible across the AI agent tooling market in mid-2026: enterprises are experimenting with agent deployment far faster than they are building the operational controls to supervise them. Industry surveys tracked by Arcade.dev have found that 88% of enterprise agent pilots never reach production, with the primary blocker cited as deployment infrastructure — isolation, governance, and compliance controls — rather than the underlying model capability. Zenity’s pitch is that it removes that specific blocker: giving security teams a way to say yes to agent deployment because the guardrails already exist, instead of blocking deployment by default.

Why Investors Are Betting on Agent-Specific Security Now

Traditional application security tooling was built to inspect code, network traffic, or user behavior — none of which map cleanly onto an AI agent that can chain together dozens of tool calls, access multiple data sources, and take actions a human reviewer never explicitly authorized. That mismatch is the gap Zenity and its rivals are racing to fill, and the size of this round — with SoftBank Vision Fund 2 and Intel Capital both writing checks — suggests investors see the agent-security category as structurally distinct from conventional cloud or endpoint security, not a feature that incumbents will simply bolt on.

1. Treat agent governance as a pre-deployment gate, not a post-incident audit

Zenity’s core technical claim — evaluating an agent’s intended action before execution rather than logging what already happened — reflects a shift enterprises need to make in how they think about AI risk. A governance layer that only flags problems after an agent has already moved funds, deleted records, or exposed data is functionally a compliance report, not a control.

2. Expect agent security spend to concentrate with cross-framework platforms

Because Zenity’s platform spans Copilot, ChatGPT Enterprise, Gemini, and Claude rather than locking into a single vendor’s ecosystem, enterprises running multi-vendor AI stacks — increasingly the norm as companies avoid single-provider lock-in — will gravitate toward governance tools that don’t require a separate control plane per AI provider.

3. Budget for agent security as a distinct line item, not a subset of existing AppSec spend

With Zenity’s revenue tripling for two consecutive years and its round attracting sovereign-adjacent capital like SoftBank Vision Fund 2, the market signal is that agent security is being treated as its own procurement category by security leaders — not an add-on module bundled into an existing security-monitoring contract.

What the Round Signals for the Broader Agent Economy

Zenity’s raise lands at a moment when the agentic AI market is expanding on two fronts simultaneously: enterprises deploying more autonomous agents into production workflows, and a parallel wave of new security vendors racing to govern them before something goes wrong at scale. The fact that a security specialist — not a foundation-model lab or an agent-orchestration platform — is pulling in a $125 million round with this investor roster suggests the market has already concluded that agent capability and agent safety are being built by different companies, on different timelines. For enterprises still in the pilot phase, that split matters: choosing an agent framework and choosing a governance layer are becoming two separate procurement decisions, not one bundled purchase, and Zenity’s bet is that most large-enterprise buyers will make the governance decision independently of which model vendor they pick.

The $180 million Zenity has now raised since its 2021 founding also says something about how investors are pricing the risk of getting agent security wrong. A breach caused by a single misconfigured cloud bucket is a bad quarter; an autonomous agent that was tricked into exfiltrating a customer database or approving a fraudulent payment because no system checked its intent before it acted is a different category of failure — one that scales with exactly the automation enterprises are trying to buy. Backers like SoftBank Vision Fund 2 and Intel Capital are not simply betting that Zenity will keep growing revenue three-fold each year, as fintech.global’s reporting indicates it has for two consecutive years; they are betting that the enterprises now piloting agents at record pace will treat governance spend as non-negotiable once the first high-profile agent-caused incident makes headlines, the same way ransomware insurance became mandatory after a wave of costly breaches rather than a nice-to-have line item.

Frequently Asked Questions

Sources & Further Reading

• Zenity Raises $125 Million in Series C Funding — SecurityWeek
• Zenity Lands $125m as AI Agent Security Race Heats Up — fintech.global
• State of AI Agents 2026: 5 Enterprise Trends — Arcade.dev