ARGORIX

AI Security SpA is the legal entity behind Governance AI.
Gobierna Tu IA is the public campaign and product communication layer.
Runtime Guardrails & Agents is the operational security and runtime control line from which this guardrail work is developed.

What is ARGORIX

ARGORIX is a SaaS platform that enables organizations to discover, control, secure, and govern the real use of artificial intelligence across their operations.

Technical definition:

ARGORIX is a multi-tenant AI governance and AI security platform that provides asset inventory, runtime guardrails, Shadow AI detection, policy enforcement, telemetry, evidence management, and operational control across applications, models, agents, and AI pipelines.

What the platform does

1. AI Inventory

ARGORIX helps organizations identify and maintain visibility over:

• AI models in use
• datasets
• prompts
• repositories
• pipelines
• AI-enabled applications
• agent workflows
• undeclared AI assets

This includes automated discovery and operational inventory.

2. Shadow AI Discovery

The platform is designed to detect unauthorized or ungoverned use of AI, including:

• undeclared repositories
• hidden AI pipelines
• unknown model usage
• shadow applications
• unregistered prompts and workflows

This is a core capability, not an optional add-on.

3. Runtime Guardrails

ARGORIX provides runtime controls to reduce operational risk in live AI environments.

This includes:

• prompt risk classification
• jailbreak and prompt injection detection
• runtime guardrails
• agent guardrails
• policy enforcement
• tool-call control
• decision validation
• traceable action logging

4. AI Risk Assessment

The platform supports continuous identification and prioritization of AI risk through:

• findings generation
• posture scoring
• remediation recommendations
• risk visibility by asset, workflow, or business unit

5. Executive Visibility

ARGORIX provides a governance layer for operational and executive audiences through:

• executive dashboards
• telemetry views
• risk metrics
• audit-oriented summaries
• governance reporting

6. AI Red Team

ARGORIX includes offensive validation capabilities for AI systems, including:

• malicious prompt simulation
• adversarial testing
• red team scenarios
• runtime behavior validation
• abuse case discovery

7. Evidence Hub

The platform is designed to support auditability and compliance readiness with:

• evidence storage
• decision traceability
• audit packs
• report generation
• investigation support

Why this matters

ARGORIX is not just documentation, policy, or compliance theater.

It is built as an operational control layer for AI.

Most organizations today can produce:

• policies
• inventories on paper
• dashboards
• governance documents

Far fewer can actually enforce control in runtime.

ARGORIX focuses on:

• real-time control
• active detection
• operational enforcement
• auditability
• security posture for AI systems

Positioning

Simple

ARGORIX helps companies use artificial intelligence without losing control over what it does.

Executive

ARGORIX enables organizations to discover, control, and audit the use of artificial intelligence across their operations.

Technical

ARGORIX is a multi-tenant SaaS platform that combines AI discovery, runtime guardrails, policy enforcement, telemetry, evidence management, and continuous governance controls for applications, models, and pipelines.

Strategic

ARGORIX is building the operational control layer that enterprises need to scale AI with security, traceability, and runtime governance.

Product lines

Gobierna Tu IA

Gobierna Tu IA is the public campaign and communication initiative that promotes responsible, secure, and governable AI adoption.

Runtime Guardrails & Agents

This line focuses on runtime security and operational controls for AI systems, including:

• prompt injection detection
• jailbreak detection
• guardrails for agents
• runtime enforcement
• policy-aware execution
• traceable AI actions

Public repositories in this organization

This organization hosts work related to:

• Spanish guardrail datasets
• QLoRA guardrail adapters
• runtime defense research
• red-team-oriented evaluation assets
• AI governance and operational control components

Creator

Gustavo Venegas
AI Security Research | AI Security SpA Chile | ARGORIX Founder & CEO

• LinkedIn: https://www.linkedin.com/in/gvenegascc/
• Research papers: https://zenodo.org/records/18918455

Legal and organizational note

AI Security SpA is the legal company behind this work.
ARGORIX is the SaaS product platform.
Gobierna Tu IA is the campaign and public-facing initiative used to communicate, promote, and operationalize AI governance adoption.

Focus areas

• AI Governance
• AI Security
• Runtime Guardrails
• Agent Security
• Shadow AI Detection
• AI Asset Discovery
• Policy Enforcement
• AI Auditability
• Evidence Management
• Red Team for AI

Mission

To give organizations the ability to adopt AI at scale with real operational control, security, and traceability.

Contact

For collaboration, partnerships, enterprise discussions, or research:

Gustavo Venegas
AI Security Research | AI Security SpA Chile Co Funder
https://www.linkedin.com/in/gvenegascc/