Works from CSV or workbook exports produced from SAP, IBM Maximo, Oracle ERP, Hexagon EAM, Infor, and other ERP, EAM, or CMMS systems. No direct integration or write-back is required — Review data requirements →
Trust Overview

AI2COE Trust Overview for diagnostic buyers.

A concise entry point for no ERP write-back, source-file handling, human review, confidence tiers, and the path to the full Trust Center.

Open Full Trust CenterReview Data Handling
Read-onlyDiagnostic workflow
No write-backSource systems remain buyer-controlled
Human reviewRequired before action
Executive AI governance layer showing risk controls, evidence records, approval checkpoints, and audit-ready decision support.
Trust controls make source traceability, confidence tiering, human review, and auditability visible before remediation decisions.
Enterprise trust posture

Proof controls buyers expect before they upload operational data.

Source purge Uploaded source files are purged after report generation; summary metrics and Open Findings remain.
No ERP write-back The diagnostic creates evidence for review. It never changes, deletes, merges, or overwrites ERP records.
Local currency Reports display money in the user's selected or country-derived currency, while USD remains the base audit calculation.
Audit trail Report ownership, access, quota, and feedback events are retained for governed review.
Session downloads Excel, Word, PDF, and CSV downloads are available only in the active generation session.
Open Findings Browser findings remain available without retaining the original source catalog rows.
Trust entry point

Start here when you need the operating boundary before upload.

AI2COE uses exported operational data to generate diagnostic evidence. The trust overview summarizes the controls buyers usually need first, then routes security, legal, procurement, and data-governance reviewers to the deeper Trust Center documents.

No ERP write-backDiagnostics do not autonomously change SAP, Maximo, Oracle, CMMS, EAM, procurement, or inventory systems.
Source-file handlingUploaded source files generate the report pack and are purged after report generation where disclosed handling applies.
Human reviewFindings remain confidence-tiered and owner-reviewed before action.
Data lifecycle

From approved export to governed evidence and source-file purge.

This overview shows the trust boundary. Open the Trust Center for control details, diligence documents, and roadmap disclosures.

Open Full Trust Center
01 Upload export Approved CSV or workbook enters the diagnostic session.
02 Run diagnostic Field mapping, source-fit checks, and selected engine analysis.
03 Generate evidence Findings, scores, confidence tiers, limitations, and owner actions.
04 Purge source file Source files are purged after report generation according to the disclosed handling boundary.
05 Retain governance metadata Open Findings, report ownership, summary metrics, usage, feedback, and audit events may remain.
Where to go next

Use the right trust page for the buyer question.

Published review document

Security Brief

Concise controls and architecture review for technical buyers.

Published review document

Data Handling Commitment

Source-file handling, purge, retained metadata, and ownership boundaries.

Published review document

DPA Template

Published data-processing terms for legal review and countersignature.

Published review document

Engagement SLA

Delivery, support, and engagement boundaries for a scoped diagnostic.

Published review document

Architecture Overview

Read-only data flow, isolation boundaries, and purge pathway.

Published review document

Procurement FAQ

Commercial and procurement questions without unsupported certification claims.

Full diligence path

The Trust Center carries the detailed enterprise review record.

Use /trust-center for implemented controls, current-versus-roadmap tables, published diligence documents, and procurement/security review paths.

Open Trust CenterRequest Security Review
Agentic AI control status

Current controls and unsupported capabilities are separated explicitly.

Tool-using AI requires identity, authority, data scope, source traceability, prompt-injection resistance, human review, evaluation, monitoring, incident handling, and decommissioning controls.

Review the risk model
ImplementedNo operational write-back and human review before action
Controlled demoFixed read-only skill allowlist, R0-R2 authority, source references, injection quarantine, and formula neutralization
Documented processIdentity, tenant, policy, incident, version, evaluation, and decommissioning requirements
PlannedShort-lived external credentials and narrowly authorized read-only connectors
Not currently supportedOperational execution, transaction submission, autonomous remediation, live OT control, or external-agent collaboration
AI2COE Copilot