(NetSentries: San Jose, CA) -- NetSentries has announced the general availability of AI-augmented security-testing capabilities throughout its expert-led targeted security assessments and within the adversarial exposure validation (AEV) workflow of its NST Assure CTEM platform.

These capabilities apply AI-assisted analysis within a strictly governed human-in-the-loop (HITL) operating model, accelerating validation of real-world exposures while preserving expert oversight, accountability, and customer control.

As part of this release, NetSentries applies commercially available frontier AI models, including various Claude models, within controlled security assessment environments to enhance reasoning, scale, exploit development, and coverage. All scoping, validation, severity determination, and reporting decisions remain the responsibility of NetSentries security professionals.

Recent advances in AI reasoning and automation have increased the speed at which weaknesses can be analyzed in complex environments. NetSentries is applying these advances responsibly to help organizations identify, validate, and remediate exploitable exposures ahead of increasingly automated and AI-assisted threats.

Arun Thomas, CTO and co-founder of NetSentries, says, “By applying AI-assisted analysis within NST Assure’s AEV workflows and our targeted security assessments, while retaining full HITL governance, we now validate exposures more efficiently without compromising safety or accountability. This ensures findings remain auditable, defensible, and actionable.”

What this means for customers

NST Assure already applies AI-assisted techniques across exposure discovery, vulnerability analysis, prioritization, and vulnerability intelligence. This new AI-assisted HITL testing enhancement within the AEV module enables controlled validation of selected exposures while preserving a human-in-the-loop delivery model. Customers receive faster confirmation on whether a vulnerability can be exploited and clearer advice on what to fix first, based on real risks, with early use showing significant reductions in validation time while still keeping oversight and the ability to audit.

Starting May 15, 2026, AI-augmented workflows are available for targeted, external zero-knowledge security assessments, subject to customer approval and engagement scoping. Credentialed and gray-box assessments will follow under the same governance model and with explicit customer consent. Final findings are issued only after human validation and quality assurance review.

NetSentries is also applying AI-assisted techniques to strengthen secure code review and application assurance, supporting the identification of complex logic flaws and insecure patterns while keeping remediation guidance expert-led and context-aware.

Human expertise is amplified rather than replaced

Every AI-assisted exploitation validation runs in semiautonomous mode with an HITL process. Expert security assessors from the NetSentries security assessment team scope the engagement, validate every finding, and own all recommendations that appear in the customer deliverables.

Controlled evolution of AI-augmented capabilities

The NST Assure platform is designed to continuously absorb additional AI-assisted capabilities under strict HITL controls. Future enhancements include:
• Adoption of newer frontier reasoning models as they mature
• Security-specialized AI models, including Mythos-class frontier systems when commercially available, to assist advanced code and vulnerability analysis within governed testing environments
• Multi-agent orchestration engines, which are needed to scale the continuous assessment of AEVs under HITL governance
• AI-augmented credentialed and gray-box assessments, enabled under explicit customer consent

Data and governance

Customer data are not used to train any AI model. All AI-assisted activity is logged and governed under internal controls, including access boundaries, use limits, and acceptable use enforcement.