Changelog

Follow new updates and improvements to Voltsec.io.

December 9th, 2025

New

Your security stack just leveled up.

We are excited to announce one of the biggest upgrades to VoltSec.io β€” the release of our AI Pentest Module (Beta).
This update brings deep learning, adaptive intelligence, and advanced adversarial testing directly into your dashboard.


πŸ”₯ What’s New in AI Pentest (Beta)

1. New Pentest Type: AI-Powered Security Testing

You can now perform pentests driven by our neural engine. This allows VoltSec to understand complex patterns, detect zero-days, analyze behaviour, and simulate adversarial attacks automatically.

2. New Test Configs Panel

A new, streamlined section allows users to configure model testing inputs such as:

  • Model Type (LLM, RNN, CNN, GAN)

  • Custom attack preferences

3. Five New AI Testing Modes

Choose the type of intelligence you want to test with:

  • Basic Testing – Standard vulnerability checks powered by AI

  • Adversarial Testing – Simulates real-world attacker strategies

  • Robustness Testing – Stress tests your models & logic against failures

  • Security Testing – AI-driven coverage for hidden security issues

  • Ethical Bias Testing – Checks for unsafe model behaviour & bias

4. Additional AI Safety Layers

We’ve added new optional modules:

  • Content Resilience Testing

  • PrismProbe Testing (model hallucination & misdirection detection)

  • Evasion Simulation (bypass attempts against your system)

5. Performance Improvements

  • Faster AI inference

  • Better accuracy in security analysis

  • Enhanced Neural Engine output

  • Optimized scanning pipeline

  • Reduced false positives

6. One-Click CEH Verified Report Generation

After your AI pentest completes, download a fully verified report β€” combining:

  • AI Findings

  • CEH Manual Verification

  • Compliance-grade summaries

πŸš€ Start Testing Smarter

The AI Pentest module is now live for all premium users with 15 Voltcharge credits.

This is just the beginning.
More autonomous attack simulations, deeper business logic analysis, and zero-day prediction models are here.