On March 17, 2026, two cybersecurity startups emerged from stealth within hours of each other. Together, they raised $182 million to solve the same problem: enterprise security operations can’t keep up with AI agents.
Kai launched with $125 million to build an autonomous cybersecurity platform. Surf AI launched with $57 million for agentic security operations. Both are betting that the entire security stack needs to be rebuilt for a world where AI agents operate inside corporate networks at machine speed.
This isn’t coincidence. It’s a market signal.
Kai: The $125M Rebuild
Kai isn’t an incremental security tool. It’s a full platform rebuilt from first principles.
The company was founded by Galina Antova (co-founder of Claroty, valued at $3 billion in industrial cybersecurity) and Damiano Bolzoni (co-founder of SecurityMatters, acquired by Forescout). Both built their careers at the IT/OT security convergence point. Now they’re targeting a bigger convergence: unifying cybersecurity, engineering, and enterprise IT into a single agentic pipeline.
The pitch: autonomous AI agents that continuously analyze threats, manage exposures, and execute response — no human bottlenecks, no siloed tools, no dashboard fatigue.
“Cybersecurity is quickly becoming a contest between AI systems,” Antova said. “The decisive factor is which side has richer data and can act at machine speed with human expert accuracy.”
Key details:
- 10 months from first line of code to launch — unusually fast for enterprise security
- Seven-figure bookings already signed across energy, pharmaceuticals, automotive, and hospitality
- Graduated from Chevron Technology Ventures Catalyst Program
- Built by PhD and postdoctoral AI researchers
- Backed by Evolution Equity Partners
The ambition extends beyond security. Kai wants to become “the AI-powered operating system that unites every cybersecurity and IT function” — the security analog of what OpenClaw is for personal agents.
Surf AI: The $57M Context Graph
Surf AI approaches the problem differently. Where Kai rebuilds the stack, Surf AI connects the existing one.
The platform ingests signals from identity, security, cloud, data, HR, and IT tools to build a context graph that maps assets, owners, permissions, and dependencies. Then it uses specialized AI agents to prioritize risks based on business impact and coordinate remediation.
“Proactive security hygiene is exactly what we’re encouraging,” said CEO Yair Grindlinger. “Surf AI connects the context, drives the remediation, and keeps your team in control at every step.”
Key details:
- $57M led by Accel, with Cyberstarts and Boldstart Ventures
- Founded by five Israeli cybersecurity veterans in 2024
- Already working with Fortune 500 companies
- Focus on closing “exposure gaps that teams have always known about but didn’t have the time or resources to address”
The context graph approach is notable because it addresses one of the hardest problems in agent security: knowing which agents exist and what they’re doing. Without inventory and context, runtime monitoring is blind.
The Pattern: Security Is Following the Agent Curve
The timing of these launches isn’t random. Look at what happened in the past 30 days:
| Date | Company | What |
|---|---|---|
| Mar 9 | AvePoint AgentPulse | Shadow agent governance |
| Mar 15 | Singulr Agent Pulse | Runtime agent governance |
| Mar 15 | Mimecast | Adaptive agent security |
| Mar 17 | Chainguard Agent Skills | Supply chain hardening |
| Mar 17 | Nvidia OpenShell | Runtime sandbox |
| Mar 17 | Kai | Autonomous security platform |
| Mar 17 | Surf AI | Agentic security operations |
Seven companies in eight days, all building security infrastructure specifically for AI agents. This is what a market category forming in real time looks like.
Why It Matters for OpenClaw Users
The enterprise security market moving this aggressively toward agent-aware tools has two implications:
1. Validation. When VCs pour $182M into agent security in a single day, they’re betting AI agents in enterprises are inevitable, not experimental. That confidence trickles down to every IT leader evaluating whether to deploy OpenClaw-style agents internally.
2. The security gap is real but closeable. The criticism that agents are “insecure by default” is valid. But the response from the security industry isn’t “don’t deploy agents.” It’s “deploy agents with proper security infrastructure.” Every one of these products exists because the market decided agents are worth securing, not worth avoiding.