Autonomous Offensive Security Built for the AI-native SDLC

RunSybil's autonomous system of agents finds exploitable vulnerabilities, verifies which ones are real, and helps teams fix them, with continuous visibility into the attack surface across the AI-native SDLC.

Point-in-time pentests, scanners, and bug bounty programs all work at human speed and stop at a single moment. Sybil detects and tests on every change.

Sybil simulates adversarial behavior to identify weaknesses before exploitation, compressing assessment timelines and enabling AppSec and engineering teams to remediate issues earlier in the development cycle.

The platform instantly retests any patch to confirm a fix actually holds, without waiting for the next scheduled engagement. This removes the operational drag of manual validation.

The RunSybil platform integrates directly into existing security workflows, feeding real-time simulation results into common defensive tools and DevSecOps pipelines.

This ensures that newly introduced risks, particularly those emerging during rapid development, are detected and surfaced without requiring disruptive, standalone testing engagements. That means complete coverage and zero exposure time, without slowing down how fast the team ships.

RunSybil came out of stealth in early 2026 with $40 million in funding. The company is based in San Francisco and was founded by CEO Ari Herbert-Voss and CTO Vlad Ionescu. Investment was led by Khosla Ventures, with participation from S32, the Anthology Fund from Anthropic and Menlo Ventures, Conviction, and Elad Gil.