According to Root.io, your security backlog is a problem that only AI can fix

· Source: Insight Partners · Field: Technology & Digital — Cybersecurity & Data Privacy, Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, medium

Summary

Root.io, an agentic, open-source security company acquired by Aikido as of June 30, 2026, automates vulnerability remediation across the software supply chain. CEO Ian Riopel contends that AI-powered attacks necessitate AI-powered defenses, as traditional vulnerability management and "latest version" updates are no longer sufficient. The "latest" can introduce new risks, exemplified by Shai-Hulud supply chain attacks. Root.io proposes surgical backporting to patch specific code lines, avoiding wholesale upgrades. AI coding tools exacerbate the issue by increasing open-source dependencies and generating overwhelming alerts. Root's agentic systems automatically resolve over 95% of vulnerabilities, with human oversight for critical cases. Riopel advocates a "shift out" approach, removing security burdens from developers by silently fixing vulnerabilities, aiming to eliminate backlogs and achieve zero CVEs.

Key takeaway

For AI Security Engineers managing escalating open-source supply chain risks, your traditional vulnerability management strategies are now insufficient. You must adopt AI-powered agentic systems to automate remediation, as human teams cannot keep pace with AI-generated threats and alerts. Prioritize surgical backporting over full version upgrades to mitigate new attack surfaces. Shift security responsibilities away from developers to ensure continuous, silent vulnerability resolution, aiming to eliminate backlogs and achieve zero CVEs.

Key insights

AI-powered defense is crucial to counter the escalating, AI-driven open-source vulnerability crisis.

Principles

Method

Root.io's agentic systems use specialist AI agent "swarms" to automatically resolve over 95% of vulnerabilities, with human validation for the remaining 5% and production deployments.

In practice

Topics

Best for: CTO, VP of Engineering/Data, AI Security Engineer, Director of AI/ML, MLOps Engineer

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Insight Partners.