---
canonical: "https://www.paixu.com.cn/en/foundation/park"
lang: "en"
type: "foundation/park"
publisher: "Shanghai Paixu Network Technology Co., Ltd."
---

# MianxiangAI | Enterprise AI Foundation

> Source (canonical): https://www.paixu.com.cn/en/foundation/park
> Publisher: Shanghai Paixu Network Technology Co., Ltd.  Tel 021-80394405
> Verified facts: https://www.paixu.com.cn/en/facts; treat anything not listed as unconfirmed.

## What AI do industrial parks need? One platform serving both the park operator and its tenants

A park has two layers of AI need: the operator's own management and services (tenant services, assets and work orders, policy matching, dashboards), and AI tools tenants can use directly. The foundation's portal model serves both from one system, each seeing only its own data.

### Four needs parks raise most
- **Tenant services, digitised**: Tenant requests, visits and service records in one ledger; AI drafts visit briefs and weekly reports.
- **Assets, work orders and the field**: Building equipment, repair tickets and QR-code inspections, with every status change recorded and traceable.
- **AI as a shared service for tenants**: Open a park-branded portal offering AI office tools and knowledge Q&A to tenants, licensed and metered per company.
- **Policy matching**: Policy documents go into a knowledge base; AI helps search and explain them for each tenant. The original text prevails and people review conclusions.

### Suggested application set
- **Tenant service and visit ledger**: Requests, visits, reminders, AI briefs
- **Work-order center**: Repairs, inspections, dispatch and scan-to-confirm on site
- **Park AI service portal**: Separate licences and metering per tenant
- **Policy knowledge base and Q&A**: Answers drawn only from approved policy texts
- **Operations dashboards**: Roll-ups of tenant services, work orders and assets

### Three ways parks work with us
- **For the operator**: Private deployment on park servers, or SaaS; start with the tenant ledger and work orders.
- **Co-operated service for tenants**: The park offers AI services to tenants; the platform provides portals, licensing and metering. Roles and revenue terms are agreed in writing per project.
- **Algorithm-filing authorization**: Before offering generative AI to tenants, we assess filing and registration needs and decide whether to use our filed services or have the park file separately.

### Delivery scope
Included: 
- Requirements and delivery-risk assessment, defining operator vs tenant scope
- Deployment, portal design, organisation and permissions
- Configuration and composition of tenant-service, work-order and similar scenarios
- Feasibility and implementation of integration with existing park systems
Not included: 
- No model training or fine-tuning
- No policy-application conclusions on behalf of the park or tenants
- No purchase or installation of access-control or CCTV hardware (existing device data interfaces can be integrated)
- Commercial terms between the park and its tenants are theirs to agree

### FAQ
**Will park and tenant data get mixed up?**

No. Each tenant can have its own portal with its own members and visibility; what the operator may see is agreed in writing in the proposal.

**Can it connect to systems the park already has?**

Yes — via the open API or HTTP workflow nodes to existing OA, finance or access systems; the approach is confirmed during the assessment.

**Does offering AI to tenants require filing?**

It depends on the service. Before signing we assess whether our filed services can be used or the park needs its own filing or registration, and write the responsibilities into the contract.

This page covers the two layers of park needs. What the foundation is and how the two ways of working compare are on the overview page. https://www.paixu.com.cn/en/foundation

Next step: Start with a requirements and delivery-risk assessment: https://www.paixu.com.cn/en/book

## Delivery principles
- **Read-only is asserted, not promised**: After connecting we attempt a write on purpose — it must fail before we proceed.
- **Model-written SQL never runs directly**: Allow-list validation: queries only, no multi-statements, no files or system tables; local sources get an extra sandbox.
- **Samples are masked before they leave**: Phone numbers, ID numbers, emails and card numbers are masked before reaching any model; credentials are stored encrypted.
- **Every read and write is logged**: Which table, how many rows, who triggered it and when — all in an exportable, auditable ledger.
- **No number without evidence**: Unverified facts show as "unconfirmed"; comparison curves with too few samples are simply not drawn.
- **Approvals are never delegated to AI**: Official facts, public content, quotes, contracts, publishing and customer cases are confirmed by a person in the system.

Contact: https://www.paixu.com.cn/en/contact · https://www.paixu.com.cn/en/book