AI and machine learning
Models trained from scratch, self-improving reasoning engines, evolutionary architecture search, plus the shipped products that lean on them.
MXCyberSLaM
An infinite, self-playing cyber range.
Active Cyber range and security training
MXCyberSLaM is a procedurally generated, verifiable cyber range paired with a tiered red/blue agent that learns to attack and defend through self-play.
What it does
The range is three things at once: the training-data factory, the reinforcement-learning reward signal, and the benchmark. Because scenarios are generated and verifiable, a solved scenario is provably solved rather than graded by a model.
It also ships a browser terminal where you practise recon, exploit, pivot and loot against live-flavoured scenarios inside a safe simulation.
Who it is for
Security teams, trainers, and researchers working on autonomous offence and defence.
What it costs
In active development; contact us about access.
Terminology
Words this page uses in a specific sense.
- Cyber range
- An isolated simulated network used to practise attack and defence without touching production systems.
- Self-play
- Training where an agent improves by competing against copies of itself — here, a red agent and a blue agent co-evolving over generated scenarios.
- Verifiable scenario
- A generated environment whose solution condition is machine-checkable, so success is measured rather than judged.
Facts
| Field | Value |
|---|---|
| Name | MXCyberSLaM |
| Status | Active |
| Category | Cyber range and security training |
| Platforms | Web |
| Pricing | In active development; contact us about access. |
| Source | Closed source |
| Publisher | Cognitive Industries |
| Published from | Brisbane, Queensland, Australia |
Capabilities behind it
Security and cryptography
Non-LLM vulnerability scanning, a self-playing cyber range, encrypted-at-rest stores, and licensing backed by proof-of-work.
Related
Last reviewed · Site changelog