Parinita Provenance
Platform-Wide Lineage Intelligence
Prove everything that happened. Across every product. On demand.
Provenance is the platform-wide chain-of-custody and lineage intelligence product that sits as the roof above all other Parinita software products. While Chrysalis acts as the immutable foundation recording every event, Provenance ties all verified events together into a single, queryable lineage graph. Its purpose: prove everything that happened.
AI Compliance Audit Prep Takes Weeks and Proves Nothing
Compliance officers, CISOs, and auditors spend weeks assembling evidence packages from fragmented logs across multiple systems. Those logs are mutable, system-specific, and not independently verifiable. For AI systems specifically, no tool answers the question: prove exactly what happened, which model was used, what data was retrieved, and who authorized it.
Regulators and auditors increasingly require AI output attestation — a defensible record of which model version produced an output, what reasoning chain was applied, and what data sources were used. No current production system provides this at the platform level.
-
Manual Evidence Assembly
Compliance officers spend weeks assembling evidence packages from fragmented logs across multiple systems — each log mutable, each system requiring separate export.
-
No AI Output Attestation
No current production system provides a legally admissible signed certificate of which model version produced an AI output, what reasoning chain was applied, and what data sources were used.
-
Provenance Gaps Before Audit Findings
Compliance gaps in a chain of custody are typically discovered during an audit examination — after the fact. No tool identifies them proactively before they become findings.
Lineage graph. Evidence packages. AI output attestation.
Provenance does not replace your compliance tooling. It is the lineage intelligence layer above Chrysalis — transforming immutable event records into queryable proof for auditors, regulators, and compliance officers.
The Investigator agent, running on Planes 1 and 3, handles supply chain and data provenance auditing across the sovereign fabric. Every query produces a Chrysalis-signed result. Every evidence package is auto-generated, not manually assembled.
-
Unified Lineage Graph
Trace any AI output back to its exact origin: model version, data retrieved, routing decisions made by Orchestra, access events from Witness, and consent grants from Sovereign.
-
Cross-Product Audit Trail
Timestamp-search events across the entire ecosystem: Central marketplace transactions, Secure threat detections, Nexus collaboration sessions, Chrysalis smart-contract executions.
-
Regulatory Evidence Packages
Auto-generate on-demand Chrysalis-signed compliance evidence for HIPAA, SOC 2 Type II, CMMC Level 2/3, FedRAMP, SEC recordkeeping, FINRA supervision, GDPR, and EU AI Act.
-
AI Output Attestation
Legally admissible signed certificate for any AI output: model version, VRC hash, data sources, inference location. Court-admissible AI provenance.
-
Security Token Provenance
SEC/FINRA-grade chain of custody for digital securities on the Chrysalis Token Engine: token ownership, accreditations, and dividend distributions.
-
Provenance Score
AI-driven risk and trust score that identifies provenance gaps in a chain of custody before they become audit findings.
-
Investigator Agent
Dedicated platform AI agent running on Planes 1 and 3, handling supply chain and data provenance auditing across the sovereign fabric.
-
On-Demand Evidence Generation
Compliance evidence packages generated on demand in minutes, not weeks. Every package Chrysalis-signed and independently verifiable.
-
Digital Securities Transfer Agent
Architected as a transfer-agent layer for digital securities issued on the Chrysalis Token Engine — full SEC/FINRA-grade chain of custody.
Six Capabilities
Unified Lineage Graph: trace any AI output to its exact origin — model version, data retrieved, routing decisions, access events, consent grants. Cross-Product Audit Trail: timestamp-search events across the entire ecosystem including Central marketplace transactions, Secure threat detections, Nexus collaboration sessions, and Chrysalis smart-contract executions.
Regulatory Evidence Packages: auto-generate on-demand Chrysalis-signed compliance evidence for HIPAA, SOC 2 Type II, CMMC Level 2/3, FedRAMP, SEC recordkeeping, FINRA supervision, GDPR, and EU AI Act. AI Output Attestation: legally admissible signed certificate detailing model version, VRC hash, data sources, and inference location.
Investigator agent. Chrysalis-sourced. Signed results.
Every Provenance query is handled by the Investigator agent on Planes 1 and 3. Investigator traverses the Chrysalis event graph, applies the compliance framework filter, and returns a signed lineage result or evidence package — independently verifiable on Chrysalis.
- 01Lineage Query ReceivedCompliance officer, auditor, or automated system submits a provenance query — AI output, timeframe, entity, or compliance framework.
- 02Investigator Agent InvokedInvestigator agent (Planes 1 and 3) begins traversal of the Chrysalis event graph.
- 03Event Graph TraversalInvestigator traces all events connected to the query: model invocations, data retrievals, routing decisions, access events, consent grants.
- 04Framework Filter AppliedEvents filtered and mapped to the requested compliance framework (HIPAA, SOC 2, FedRAMP, etc.).
- 05Gap DetectionProvenance Score engine identifies gaps in the chain of custody — events missing Chrysalis anchoring, incomplete consent chains, jurisdiction violations.
- 06Evidence Package ConstructionChrysalis-signed evidence package constructed from the filtered event graph. Formatted for the requested regulatory framework.
- 07Attestation Certificate IssuedFor AI outputs: legally admissible signed certificate issued with model version, VRC hash, data sources, and inference location.
- 08Result DeliverySigned lineage graph, evidence package, and Provenance Score returned to requester. Independently verifiable on Chrysalis.
# Query AI output provenance via Investigator:
from parinita import provenance
result = provenance.Investigator().query(
output_hash='sha256:abc123...',
framework='hipaa',
include_attestation=True
)
print(result.lineage_graph)
print(result.evidence_package) # Chrysalis-signed, HIPAA-formatted
print(result.provenance_score) # Gap identification before auditProven at scale. Not in a lab.
Parinita AI Edge is the production deployment of the Parinita platform and the largest heterogeneous AI infrastructure deployment in the United States.
Parinita AI Edge
The most complex heterogeneous AI infrastructure in the United States. 101 sites, 9 planes, 12,000+ nodes, 4 accelerator vendors, dual network fabrics, four-layer tenant isolation — all through a single sovereign control plane.
Network & Security Infrastructure
- Multi-vendor acceleratorsFour accelerator vendors — Intel Habana, NVIDIA, AMD, Qualcomm — orchestrated through one control plane with unified scheduling, monitoring, and lifecycle management.
- Dual-fabric networkingCisco production fabric and Arista GPU backend fabric operating as a coordinated system, bridged by identity-aware routing.
- Nationwide scale101 sites across 42 U.S. states, each operating autonomously with a local control agent and a sovereign cross-site routing plane.
- Multi-tenant isolationFour-layer defense-in-depth: VXLAN VNIs, identity-routing, Palo Alto firewalls, and Cilium eBPF — validated across every plane and site.
- Compliance readinessFIPS 140-2 at launch, with FedRAMP Moderate, CJIS, and IL4/IL5 certification paths active through Parinita compliance profiles.
- Sub-millisecond routingEvery request classified and dispatched in under 1ms, enabling real-time SLA enforcement without perceptible overhead.
Built for compliance officers, CISOs, and auditors.
Provenance does not ask you to change your compliance processes. It is the query layer above Chrysalis — answering prove everything that happened without manual evidence assembly.
AI compliance audits (EU AI Act, FedRAMP, CMMC), regulatory examinations (SEC, FINRA), healthcare compliance (HIPAA, FDA 21 CFR Part 11), and any workflow requiring AI output attestation with a legally admissible signed certificate.
-
AI Compliance Audits
EU AI Act, FedRAMP, CMMC Level 2/3 — provenance of every AI output across the platform, auto-generated evidence packages, AI output attestation certificates.
-
Regulatory Examinations
SEC and FINRA examinations requiring non-repudiable audit trails of financial decisions and AI-assisted recommendations. Provenance auto-generates the evidence package.
-
Healthcare AI Compliance
HIPAA and FDA 21 CFR Part 11 — certified data lineage for FDA submission, AI output attestation for clinical decision support recommendations.
-
Legal and Forensic Investigation
E-discovery and forensic investigation requiring a complete, tamper-proof chain of custody from AI output back to source data, model version, and access events.
-
Digital Securities Compliance
SEC/FINRA-grade chain of custody for digital securities on the Chrysalis Token Engine — transfer agent layer for token ownership, accreditations, and dividend distributions.
Access Models
Provenance is available to all Parinita platform users as the lineage query layer above Chrysalis. Standalone Provenance access for compliance officers and auditors is available as a separate role.
Provenance queries from any Parinita surface — lineage graph, evidence packages, Provenance Score | Best for: Platform users needing compliance evidence on demand
Dedicated Provenance access for compliance officers and auditors — framework-specific dashboards, scheduled evidence package generation | Best for: Compliance teams
Custom compliance framework integration, digital securities transfer agent layer, Investigator agent dedicated capacity | Best for: Regulated enterprises and financial institutions
Request a Demo
Our engineering team has deployed Provenance as the lineage intelligence layer across 101 sites. We bring that operational experience to every deployment conversation.