Government 2050: SSOT, once-only principle, and agentic digital services
By 2050 government transitions from a model of application-based services to a model of proactive service with AI agents, single source of truth on the citizen, and the once-only principle — data is collected once and used by all agencies by consent. Authorial position on the trajectory of digital government in Uzbekistan.
Discuss Your ChallengeAuthorial position: by 2050 government in Uzbekistan can become a platform for the citizen’s life with a single source of truth, once-only principle, and AI agents as universal interface. This requires a long-term rebuild program over 10-15 years.
How this shows up in real life
The contours of the future are already visible. Estonia with X-Road operates on the once-only principle since the 1990s. Singapore with MyInfo gives the citizen a single window to data. UK with GDS showed how government services can be redesigned around the citizen. Finland, Denmark, South Korea — other countries moving in this direction. AI agents for government services pass pilots in several countries. Uzbekistan has built visible digital government infrastructure but registry fragmentation and application-based model remain behind the facade.
Why the bank ended up here
Government systems were historically built separately by agencies as tasks emerged. Each agency — own database, data format, services. The model — citizen approaches the agency with an application, the agency processes. This worked in the paper era. In the digital era this model is outdated — the citizen expects the government to know them and approach first at the right moment. Agency fragmentation prevents fulfilling this expectation. Without a single source of truth and the once-only principle, any digitalization remains a facade.
What teams usually try — and why it does not fix it
- Creating yet another government services portal — does not solve fragmentation, adds a channel
- Procuring an MDM system — without data steward and accountability policy the system degrades
- Appointing CDO of government as accountable — without inter-agency mandate nothing happens
- Data cleansing program by order — produces point effect, discrepancies return in a year
- Launching AI agents in portals — without unified data underneath, agents give contradictory answers
- Partnerships with fintech for KYC — pointwise work but do not change the overall model
What type of solution is actually needed
By 2050 the winning government is the one that has rebuilt itself as a platform for the citizen's life. Architecturally — single source of truth (SSOT) with biometric identification, once-only principle for data, AI agents as universal interface. Operationally — small central team of architects and product people; agencies focus on their data and competence domains. Regulatorily — modern personal data protection framework with managed consents, inter-agency data sharing by citizen consent. Customer experience — the government approaches the citizen proactively through an AI assistant understanding their life context.
What to check before starting
- Are sources of truth assigned in government for key data domains
- Does an inter-agency agreement on data sharing with once-only principle exist
- Does the personal data regulator work as a real controller, not formality
- Is digital government architecture ready for AI agent integration
- Does a long-term digital government vision exist over 10-15 years, not just 3-5
How to move step by step
- Articulate the long-term digital government vision for Uzbekistan over 10-15 years with once-only and SSOT
- Get government-level mandate to rebuild the inter-agency data model
- Launch the single source of truth program — identity resolution layer, unified citizen profile, update flows from agency systems
- In parallel — update the personal data protection regulatory framework for modern scenarios
- Implement proactive services on top of the unified profile — personal companion, life event orchestration
- After 5-7 years — pilot AI agents as universal interface to government services
- After 15-20 years — full maturity of once-only model and AI-assisted services
What else is worth exploring
Topics from the same area we usually explore together
Document Management
Document management is not about scanning paper. It's about getting the right document to the right person at the right moment — without…
→SolutionAutomation
Automation is not replacing people with robots. It's eliminating steps that shouldn't exist and accelerating those that should.
→SolutionCitizen Data Platform: a single citizen master for the state
Customer master for the state. Citizens in one place, distribution to downstream agencies, consent-aware. Foundation for all digital-first…
→SolutionInter-Agency Workflow Platform
Cross-agency processes (citizen permits, business licenses, large procurements) require coordinated workflow. Platform replaces email and…
→I do not just write about this. I can come in, examine your situation and design a solution for your specific landscape.
Discuss applying this →Ready to discuss your challenge?
Tell me what's not working or what needs to be built. First conversation — no obligations.
Usually respond within a few hours