Implementation Sandbox

From Framework to Working Service

The DPI-AI Framework is designed to be implemented — not just read. Use these tools to assess your readiness, explore real deployment scenarios, generate a phased roadmap, and watch a live end-to-end demo of a farmer requesting government support via WhatsApp.

1
Assess your country's readiness across 5 dimensions
2
Watch the live farmer demo — the framework in action
3
Explore your sector's implementation journey
4
Generate a tailored 3-phase roadmap
How it works
AI Blocks provide modular, callable functions
·
DPI Workflows orchestrate the sequence
·
Public Agents deliver to citizens at scale
·
All governed, auditable, replaceable

🌾 Farmer Benefit Application — End-to-End Demo

Amina is a smallholder farmer in Kisumu, Kenya. She sends a voice message on WhatsApp to apply for the Ministry of Agriculture's support benefit. Watch how AI Blocks, DPI Workflows, and DPI systems combine to deliver the service — with full governance at every step.

Speed:
🌿
Kilimo Bot 🤖
Ministry of Agriculture · online
🎤
📱 M-Pesa Notification · Mojaloop
You have received KES 3,000 from Ministry of Agriculture. Ref: AG-2026-8847. Balance: KES 3,847.

🤖 Public Agent + DPI Workflow — live execution

🤖
public_agent.init()
PublicAgent("KilimoBot").receive(channel:"whatsapp", user:"0712xxx")
✓ Agent initialised · consent logged · workflow: AGRI_BENEFIT_2026
🎤
speech_to_text()
speech_to_text(voice_note, lang:"sw-KE")
✓ confidence: 0.94 | "Nataka msaada wa kilimo"
🌐
translate()
translate(text, source:"sw", target:"en")
✓ "I want to apply for agricultural support"
🔍
structured_extract()
extract(text, schema:{intent,name,location})
✓ intent: agricultural_support | name: Amina
🪪
identity_verify() · MOSIP
mosip.verify(id:"38291047", biometric:false)
✓ Amina Odhiambo | Kisumu | DOB: 1988-04-12
✅
eligibility_verify() · OpenSPP
openspp.check_eligibility(national_id, "AGRI_SUPPORT_2026")
✓ ELIGIBLE · Maize 2.5ha · enrolled 2024-01
📋
submit_application() · OpenSPP
openspp.submit_benefit_application(farmer_id, ref:"AG-2026-8847")
✓ Application #AG-2026-8847 · status: approved
💸
payments.disburse() · Mojaloop
mojaloop.transfer(KES_3000, msisdn:"0712xxx")
✓ KES 3,000 → M-Pesa 0712xxx · ref: MOJ-9921
📊
audit.log()
audit.log(session_id, actions[], outcome:"success")
✓ Full trail logged · consent verified · PII redacted
About this demo: This is a representative simulation of the Livelihoods use case from the DPI-AI Framework paper. The Public Agent (KilimoBot) orchestrates the full interaction. speech_to_text() and translate() run on cloud or sovereign-hosted models. identity.verify() calls MOSIP (open-source digital identity). Eligibility and benefit submission use OpenSPP (open social protection platform). Payment disbursement uses Mojaloop rails to M-Pesa. The entire interaction is governed by the DPI Workflow YAML spec — AI operates inside it, never outside it. See implementation guide →
📊

AI-DPI Readiness Assessment

Answer 16 questions across five dimensions — Technology, Governance, People, Data, and AI Adoption — to understand your country's readiness and receive tailored recommendations. Takes about 6 minutes.

Explore an Implementation Scenario

Select a use case to walk through the step-by-step implementation journey — including the AI Blocks required and governance decisions at each stage.

🛡️
Social Services
Benefits eligibility & disbursement
📝
Civil Registration
Birth certificates & identity
🌾
Livelihoods
Farmer benefits via WhatsApp
📚
Education
Personalised learning support
🚨
Crisis Response
Emergency benefit disbursement
🗺️

Generate Your Implementation Roadmap

Complete the Readiness Assessment first to get a personalised 3-phase implementation roadmap tailored to your country's gaps and strengths.

Or generate a generic roadmap:

Country Archetype Comparison

Compare your readiness profile to four common country archetypes. Complete the assessment to overlay your profile.