ConfX Global
InsightXOverviewReportsSector BriefingsLeadership PulseKnowledge SessionsInterviewsArticlesTools
Sector briefing · AI and automation

Public-sector AI: procurement and accountability

How governments are buying and overseeing AI systems.

ConfX editorial2 October 20263 min readGovernment and public sector
Download PDF

Why it matters

Public bodies are adopting AI for casework, translation and service delivery, with high expectations for transparency and fairness. Procurement is becoming a key point of control.

Public bodies must be able to explain decisions and show that they are fair and lawful. Procurement contracts are a key way to secure transparency, testing and access to information from suppliers.

By the numbers

31%of Americans trust their government to regulate AI 1
81%in Singapore, the highest 1
Trust in government to regulate AI
Singapore
81%
Global average
54%
United States
31%
Source: Stanford HAI, The 2026 AI Index Report, April 2026

What's changing

  • The EU AI Act requires public bodies deploying high-risk systems to assess the impact on fundamental rights.
  • Several countries publish algorithmic transparency records.
  • Model contract clauses help buyers set AI requirements.

How leaders are responding

  • Including transparency, audit and data-rights requirements in tenders.
  • Keeping a register of AI systems in use.
  • Building in-house expertise to evaluate suppliers.

What good looks like

  • Procurement terms that require transparency, testing and audit rights
  • A public register of AI systems in use
  • Impact assessments completed before deployment

What to watch

  • Public-sector obligations under the EU AI Act.
  • Algorithmic transparency registers.
  • Shared AI procurement frameworks.

Where to start

Create an inventory of AI tools in use, including those inside purchased software, and assess each for its impact on citizens.

Questions for your leadership team

  1. Which decisions affecting citizens involve AI?
  2. What can we audit under our contracts?
  3. How do we explain our use of AI to the public?