The AI Factory's safety floor. 60+ managed guardrails.
60+ safeguards enforced at the platform layer, with a core ten every customer gets on day one. Data protection, accountability, transparency, and oversight built in. ISO 27001 and CPS 234 evidence exported on demand.
Your engineers build the product. We keep the safety floor.
What you get
Data governance and security
Classification, encryption, and provenance tracking enforced at ingestion. Cybersecurity controls applied to every AI data flow.
Accountability process
Clear ownership, internal capability, and regulatory compliance strategy. Training programs for teams that build and operate agents.
Risk management
Stakeholder-impact assessments and ongoing risk reviews. Mitigations mapped to system usage and updated continuously.
Human control and oversight
Intervention points across the AI lifecycle. Meaningful human review for decisions that matter. Documented escalation paths.
User disclosure and challenge
End users told when they are talking to AI. Clear paths to challenge a decision, outcome, or interaction.
Record keeping and testing
Comprehensive AI inventory. Pre-deployment testing and continuous monitoring for behaviour change or unintended consequences.
Everything included. Fixed monthly fee.
Operations
- 24/7 monitoring of guardrail enforcement
- Continuous framework patching
- Policy drift detection and remediation
- Incident response for safety violations
- Monthly compliance posture review
Engineering
- Core ten guardrails deployed on day one, with 60+ total available
- Custom-policy authoring and review
- Evidence-lake schema and export pipelines
- Data-classification tooling
- Advisory on new AI regulation
Compliance
- ISO 27001 evidence pack export
- CPS 234 control mapping
- APRA CPS 234 mapping on request
- Auditor-facing read-only access
- Stakeholder-engagement records
How it works
Guardrails are the first service deployed. Agentic Ops and MLOps inherit them.
Assess
Your data, frameworks, and current controls. We map the gap to the core ten and prioritise the rest of the 60+.
Build
Platform controls deployed. Policies authored in Git. Evidence lake stood up inside your AWS account.
Operate
24/7 enforcement monitoring. Every AI decision logged. Violations surfaced before auditors ask.
Improve
New regulations tracked and absorbed. Policies iterated with your product team. Framework versions rolled forward.
Audited and certified
AWS Advanced Partner
ISO 27001 Certified
AWS SaaS Competency
See the guardrails mapped to your stack.
Send us your current compliance posture and we will show you what changes.
Frequently asked questions
Are guardrails a separate product?
No. Guardrails are the Factory's safety floor. You can still engage this scope standalone if you need guardrails first and will add agents or MLOps later.
Which frameworks do the guardrails map to?
ISO 27001 and CPS 234 out of the box. APRA CPS 234 and SOC 2 with light mapping during onboarding.
How are the guardrails enforced?
Through platform controls, not policy documents. Every AI decision logged to an evidence lake. Misuse flagged by the Security Sentinel.
Can we add our own guardrails?
Yes. 60+ are available across categories, with ten as the core baseline. Customer-specific policies sit on top in your Git.
Who owns the evidence pack?
You do. Evidence lives in your AWS account. We export it in the format your auditors ask for.
How quickly are new regulations absorbed?
Our R&D crew tracks incoming regulation and rolls new controls out as versioned changes. You inherit the update without a procurement cycle.