Project Overview
I. Core Use Case: Embedded Endpoint for Hong Kong Government Grant Advisory Services This section demonstrates how AI-powered identification capabi...
- Client
- 2DQY Project
- Project Year
- 2026
- Deliverables
- A multi-functional AI conversational platform supporting multimodal interaction and custom script extensions, delivering seamless automated communi...
Annex Chatbot
I. Core Use Case: Embedded Endpoint for Hong Kong Government Grant Advisory Services
This section demonstrates how AI-powered identification capabilities are embedded into the HKSAR Government’s grant service ecosystem, transforming complex policy documents (e.g., *Application Guidelines*, *Compliance Handbooks*) into interactive data flows.
| Use Case Category | Core Logic & Technical Implementation | | :--- | :--- | | Policy Intent Smart Routing | Provides `chat-automation` capability for platforms like *GrantConnect*. Automatically detects user identity (e.g., local SME holding a Business Registration Number) and intent to route precisely to relevant schemes — e.g., BUD Fund, TVP, or EMF. | | Eligibility Criteria ETL Pipeline | Builds an end-to-end chain: User Q&A → Eligibility Factor Extraction → Structured JSON Fields. Automatically identifies key data points (e.g., annual turnover, MPF contribution records) to pre-populate official forms from HKPC or policy bureaux. | | Dynamic Knowledge Base Interaction | Leverages real-time endpoint integration to fetch latest grant terms (e.g., 2026 updates to *e-Commerce Easy*), improving AI accuracy in automated eligibility checks and ensuring recommended schemes align with current policy. | | Cantonese Semantic Mediation Layer | Processes colloquial queries (e.g., “Any money if I start an online store?” or “I want to exhibit in mainland China”) and maps them to formal scheme names (e.g., TVP Technology Voucher, BUD Special Fund), outputting clean, standardised data.
II. Business Value: Deployment on Local Grant Service Platforms
This section addresses pain points faced by Hong Kong enterprises applying for grants, demonstrating how the platform directly empowers executing agencies such as the Hong Kong Productivity Council (HKPC), trade associations, and professional advisory teams.
1. Automated Consultation & Operational Efficiency
- Guided, Decentralised Screening: Uses branching logic (Q1–Q3) to enable users to self-screen eligibility, significantly reducing operational load on human agents handling repetitive initial queries like “Is my company eligible?”
- Multilingual Grant Navigation: Single logic layer supports seamless switching between Traditional Chinese, Simplified Chinese, and English — helping both local SMEs and foreign-invested firms quickly locate schemes such as PAG or CMDF, boosting conversion rates.
2. Professional Terminology & Compliance Governance
- Rule-First Matching Mode: Enforces predefined policy rules strictly (e.g., non-local salaried staff cannot count as R&D personnel under TVP). AI enriches responses with precise contextual detail, guaranteeing 100% accuracy in policy interpretation, eliminating misguidance.
- Closed-Loop Review Pathway: Implements a “dialogue-as-pre-assessment” mechanism. Administrators can instantly review Phase 1/2 conversation JSON logs, rapidly assessing applicant authenticity and compliance risk — reducing audit burden and downstream exposure.
3. Data Collaboration & Secondary Conversion
- Automated Lead Capture: One-click export of conversation JSON transfers users who complete preliminary eligibility checks directly into CRM systems or advisor workloads — shortening the journey from “consultation” to “formal application”.
- Real-Time Event Notification (Webhook): Triggers instant alerts to advisors when users express high-match intent (e.g., alignment with the 2026 New Industrialisation Grant) — enabling timely intervention during critical application windows.








