Whether you embed Xybern into your AI product or deploy it above all your AI systems, the governance standard is the same.
Embed Xybern into your AI product stack as a verification and enforcement layer.
Sits between your AI models and end users. No model changes required.
Every output is broken into individual, verifiable claims mapped to evidence.
Deterministic scoring with zero variance. The same input always produces the same score.
Policy enforcement, bias detection, and custom rules enforced on every output.
Every verification is permanently recorded with cryptographic hash chain integrity.
Deploy Xybern as an infrastructure layer above all AI systems.
Both models share the same verification engine and governance standard.
| Feature | Embedded | Centralised |
|---|---|---|
| Deployment | Inside your AI product | Above all AI systems |
| Verification Engine | Full engine embedded | Full engine centralised |
| Control Layer | Per-product rules | Organisation-wide rules |
| Provenance Vault | Per-product ledger | Unified ledger |
| Trust Scoring | Deterministic | Deterministic |
| Multi-Model Support | Your product models | All organisational models |
| Agent Tracing | Product agents | All agents across org |
| Behavioural Enforcement | 8-category detection | 8-category detection |
| Data Residency | Inherits product region | Workspace-level control |
| API Access | Embedded SDK | Full API + SDK |
Regardless of deployment model, every AI output goes through the same verification, governance, and audit pipeline.
Claim decomposition, evidence mapping, and deterministic scoring, identical across both models.
Regulatory templates, bias detection, and custom rules enforced with the same engine.
Immutable ledger with SHA-256 chain integrity and Merkle proof support.
Same API, different deployment context.
// Your product code
import { Xybern } from '@xybern/sdk'
const result = await xybern.verify({
content: modelOutput,
model: "gpt-4",
context: "product-chat"
})
// result.trust_score → 0.94
// result.governance → "PASS"// Organisation-wide governance
POST /api/v1/verify
{
"content": "AI output text...",
"source": "employee-copilot",
"department": "legal",
"jurisdiction": "EU"
}
// Same engine, org-wide rulesDeploy with confidence.
Embedded or centralised, deploy Xybern the way that fits your infrastructure. Same pipeline. Same enforcement. One standard.