Endor Labs Acquires Autonomous Plane to Combat AI-Generated Code Risks
The security firm is shifting toward AI-native analysis to manage the surge in code volume and vulnerability noise.
Endor Labs is pivoting toward AI-native application security to combat the risks associated with the rapid rise of AI-generated code. The company recently acquired Autonomous Plane to provide deeper visibility into how AI-produced code behaves within real-world applications.
Varun Badhwar, CEO and Co-founder of Endor Labs, detailed this strategic shift in interviews with Pulse 2.0 and CIO Influence. The acquisition of Autonomous Plane is designed to enhance the company's ability to track risks across both traditional dependencies and AI-generated code, specifically focusing on full-stack reachability from the initial code to the final container.
The Compression of Development
Modern software relies heavily on open-source software and third-party libraries, but the introduction of AI copilots and autonomous agents has fundamentally changed the production landscape. According to Badhwar, "AI is fundamentally compressing the software development lifecycle," leading to a dramatic increase in the speed and volume of code being written.
This acceleration has rendered traditional security methods obsolete. Periodic security scans and manual reviews are no longer sufficient to manage the scale of vulnerabilities and supply chain risks created when AI can generate vast amounts of code in seconds. Consequently, Badhwar argues that the industry must move away from reactive scanning in favor of continuous, intelligent security analysis.
Solving the Signal Overload
As AI-driven development becomes mainstream, the attack surface for enterprises expands rapidly. This creates a critical bottleneck for security teams who are often overwhelmed by "signal overload"—thousands of alerts that make it difficult to identify genuine threats.
To maintain development speed without sacrificing trust, Endor Labs is pushing the industry toward a model based on "reachability" and "exploitability." This approach prioritizes vulnerabilities that can actually be executed in a live environment rather than flagging every theoretical flaw. As Badhwar puts it, "The challenge isn’t finding problems—it’s figuring out which ones actually matter."
The Path to AI-Native Security
The move toward AI-native security reflects a broader industry trend where the role of the developer is evolving. Rather than writing every line of code, developers are increasingly tasked with guiding and validating the output of AI systems.
By integrating the capabilities of Autonomous Plane, Endor Labs aims to provide the visibility necessary to validate this AI output in real-time. The focus remains on ensuring that the increased velocity of the software development lifecycle does not introduce unmanageable risks into the production pipeline.