Modernise without the big bang: Governed AI as a delivery safeguard for legacy public services

2026-05-20 · Source: Thoughtworks Insights · Field: Government & Public Sector — Digital Government & E-Government, Public Policy & Governance, Public Finance & Administration · Depth: Intermediate, medium

Summary

Public sector modernization is shifting from "big-bang" rewrites to a "thin slices" approach, leveraging governed AI to mitigate strategic risks associated with legacy systems. This redefines legacy beyond technical debt to encompass operational, regulatory, and public trust risks. AI acts as a catalyst for data hygiene, exposing inconsistencies and improving quality, and functions as a human-assisted collaborator in tasks like case review and reverse-engineering legacy logic, reducing costs by around 30% in some contexts. The approach emphasizes designing for data sovereignty and equity, rigorously testing models across diverse populations, and fostering cultural adoption by automating low-value tasks. A three-step playbook involves reverse-engineering current systems, implementing small, high-value "thin slices," and then measuring and scaling.

Key takeaway

For Directors of AI/ML or IT Professionals tasked with modernizing public services, abandon "big-bang" rewrites. Instead, adopt a "thin-slice" approach using governed AI to de-risk legacy systems and improve data quality incrementally. Focus your efforts on AI-assisted human collaboration for tasks like case review or reverse-engineering legacy logic, ensuring data sovereignty and equity are embedded from the start. This strategy allows you to prove value and reduce risk in 3-6 month cycles, gaining crucial evidence for stakeholders.

Key insights

Governed AI facilitates incremental modernization of legacy public services by de-risking strategic challenges and improving data quality.

Principles

• Legacy systems pose strategic, not just technical, risks.
• AI implementation drives essential data hygiene.
• AI functions as a human-assisted collaborator.

Method

Modernize via a three-step playbook: reverse-engineer legacy systems with AI, implement a "thin slice" for a specific outcome on modern foundations, then measure impact, learn, and scale.

In practice

• Use AI for case review to triage applications.
• Employ AI to infer legacy system business rules.
• Automate tedious manual tasks, freeing staff.

Topics

• Governed AI
• Public Sector Modernization
• Legacy Systems Management
• Data Hygiene
• AI-assisted Collaboration
• Data Sovereignty
• Digital Transformation

Best for: Director of AI/ML, IT Professional, Consultant

Related on AIssential

• State AI Rollouts Are Outrunning Their Own Governance — Rapid state-level generative AI deployment often outpaces the establishment of robust governance and transparency frameworks.
• How AI-Driven Workflows Are Changing the Way Companies Think About Data Risk — AI-driven workflows transform data risk management by automating threat detection and enhancing compliance, but require robust governance.
• Establishing AI and data sovereignty in the age of autonomous systems — Enterprises are prioritizing AI and data sovereignty to mitigate IP risks associated with third-party generative AI models.
• The US Government Tech Market Enters A New Phase Of Change — AI is reshaping US government technology strategy, driving modernization and a shift to integrated capability ecosystems.
• Digital monitoring is growing in South Africa’s public service – regulation needs to catch up — Rapid digital tool adoption in public sector evaluation outpaces ethical governance, creating risks for citizens and data integrity.
• Architecting Data And AI In The Era Of Enterprise Intelligence: Meet Shylaja Nathan, Principal Analyst — Agentic AI demands modern data architectures that provide context, responsiveness, and semantic consistency at scale.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Thoughtworks Insights.