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

# MianxiangAI | Enterprise AI Foundation

> Source (canonical): https://www.paixu.com.cn/en/foundation/sme
> 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 SMEs actually need? Get one scenario working before anything else

SMEs rarely lack a model. What is missing is the step that puts AI into daily work: who answers enquiries, who keeps the ledger, who drafts quotes and documents. Start with one scenario that happens every day, run it on what the foundation already has, then decide what comes next.

### The four needs we hear most
- **No missed enquiries**: Enquiries from the website, WeChat and forms land in one lead pool; AI pre-sorts and suggests replies, people confirm and follow up.
- **AI drafts the paperwork**: Quote notes, proposals, weekly reports and first-draft clauses come from AI; people edit and sign off. Approvals stay human.
- **One ledger first**: Customers, projects, assets and expiry reminders run on ready-made ledgers — no months of requirement meetings first.
- **Be found in AI search**: Turn brand facts, FAQs and site content into a form AI can cite, and monitor how AI assistants answer questions in your field.

### Suggested starting applications
- **Lead pool with AI triage**: Multi-channel enquiries, assignment and follow-up reminders
- **AI documents, sheets and slides**: Draft, rewrite and tidy up; people finalise
- **Company knowledge Q&A**: Answers from approved content only; restricted material stays out of the shared index
- **Asset and expiry ledger**: Reminders before domains, certificates, contracts and licences expire
- **Website + GEO visibility**: Make sure AI search describes you accurately

### How SMEs usually work with us
- **Start on SaaS**: Turn on only the modules you use and prove one scenario; move to private cloud later without redoing the setup.
- **Use our filed AI services**: No need to build models; if you offer AI features to the public, we first assess the filing and registration requirements for your scenario.
- **Small, staged customisation**: When off-the-shelf is not enough, assess first and customise in stages — no all-at-once rollout.

### Delivery scope
Included: 
- Scenario mapping and a requirements and delivery-risk assessment
- SaaS activation or private deployment, with organisation, roles and permissions
- One core scenario configured and composed into production
- Training and post-launch follow-up
Not included: 
- No model training or fine-tuning
- No guarantee that AI output is always right — key conclusions need human confirmation
- No approvals or external commitments on your behalf
- Rich real-time interfaces need separate front-end work, assessed separately

### FAQ
**On a tight budget, where should we start?**

Start with something that happens every day and shows results directly — handling enquiries or drafting documents. Prove it on existing capabilities, then decide whether to expand. Do not start with the whole suite.

**Do we need our own model account?**

You can bring your own model account or use the platform default. Models can be switched at any time without redoing business configuration.

**Where does our data live?**

Your choice of SaaS or private cloud; with a private deployment the data stays on your own servers.

This page covers the SME starting point. 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