Aikido Security BV today introduced Infinite, a continuous artificial intelligence penetration solution that autonomously validates and fixes vulnerabilities.

Penetration testing, or pentesting, is the simulation of cyberattacks performed by ethical hackers or automated systems to evaluate the effectiveness of security systems, networks or defensive tools. It’s designed to mimic attacks used by real-world bad actors, using actual techniques and tools that exploit current and zero-day vulnerabilities before they can be exploited.

Traditional and manual pen testing is often done weeks after the software is shipped. This means that it has already been deployed and remains open to attack for a significant period before receiving suitable testing.

Aikido said AI-driven testing and solutions can be riddled with false positives, do not operate truly autonomously and can stop short of fixing problems. Infinite is built to close this gap.

“For years, organizations have been working around the limitations of traditional testing and dynamic application cyber security because there was nothing better available,” said co-founder and Chief Executive Willem Delbare.

According to Aikido, the fast pace of software development has opened up chinks in the armor of security where it’s difficult to stay on top of features as they’re shipped. Today’s development cycle is continuous; security testing and vulnerability repair must also be continuous.

With Infinite, each software change triggers AI pentesting agents that discover risks, validate exploitability, apply remediation where safe and retest to confirm risk reduction.

Instead of relying on the traditional periodic window, security teams can now apply penetration testing based on release cycles and use the power of automation and a feedback loop empowered by tireless agentic AI to scale their deployment.

“As software delivery cycles shorten and change becomes constant, point-in-time security models no longer reflect how modern systems operate,” said Katie Norton, research manager for DevSecOps and software supply chain security at IDC.

Aikido said its AI pentesting capability has proven able to reveal complex vulnerabilities in apps and frameworks previously undiscovered by experts. It works best in combination with in-house expertise to build better attack profiles and fix vulnerabilities to improve security posture.

Images: SiliconANGLE/Microsoft Designer, Aikido Security