Shared Standards, Stronger Public Services: Making AI Work Together

Aug 21

3 min read

In an article from Resultsense they highlighted the call towards the end of last year from digital leaders within local government and nonprofits of the need for shared frameworks, transparent governance and citizen-centred design to realise AI's promise in public services.

The commonality of challenges and the need for consistent shared approaches was highlighted by Julian Patmore, Service Director, Digital & Customer Access at Peterborough City Council, who said that “All of us are going through the same challenge around what AI means as an organisation,” and that councils are “speaking to each other and wanting to learn from each other.”

Julian Patmore

Why Common Standards Matter

Public bodies are working at different stages of AI maturity. Some are testing focused use cases; others are moving towards organisation-wide approaches. Without a common foundation, however, each organisation can be left to interpret risks, controls and assurance requirements independently.

That fragmentation can result in duplicated effort, inconsistent safeguards and difficulty scaling solutions across services or places. Common standards can provide a more reliable basis for responsible innovation by helping organisations align around:

Clear accountability for AI-supported decisions.
Strong data governance, including data quality, security and lawful use.
Transparency for residents, service users, staff and stakeholders.
Meaningful human oversight where AI may affect an individual’s rights, opportunities or access to support.
Fairness, accessibility and inclusion throughout the service lifecycle.
Evaluation methods that assess service outcomes as well as technical performance.

The UK Government’s AI Playbook similarly directs organisations to use AI safely, effectively and securely; it also links AI-enabled service development to the Government Service Standard and, for organisations in scope, the Algorithmic Transparency Recording Standard.

Collaboration Turns Principles Into Practice

Standards are most useful when they help delivery teams make sound decisions in real circumstances. This is where collaboration matters.

Local authorities, public-service providers, charities, professional bodies and suppliers all hold part of the knowledge needed to design effective AI-enabled services. Frontline teams understand operational pressures and resident needs. Service users can identify barriers and unintended consequences. Digital and data specialists bring technical expertise. Governance, legal and information-governance colleagues help ensure solutions are safe and defensible.

Datnexa believes that responsible AI programmes should create deliberate forums for these perspectives, not seek them only at the final assurance stage. Collaboration should begin when a problem is defined and continue through design, testing, deployment and ongoing monitoring.

From Fragmentation to Reuse

Shared approaches can help organisations move more quickly while maintaining appropriate control. Rather than repeatedly creating policies, assessment methods and technical safeguards from scratch, public bodies can reuse proven patterns and improve them collectively.

For example, a council developing an AI-enabled tool to support contact-centre triage might use an agreed cross-sector assessment framework to document:

The service problem and intended benefit.
The data used and its limitations.
Accessibility, equality and privacy considerations.
Where staff review, override or escalate AI-generated outputs.
How accuracy, resident outcomes and unintended impacts will be monitored.

This does not remove local accountability. It gives each organisation a clearer, more consistent way to demonstrate that it has exercised it.

Responsible AI Is a Shared Capability

The objective is not uniformity for its own sake. Public services vary, and local context matters. The objective is to establish a shared baseline that makes responsible AI adoption more repeatable, transparent and trustworthy.

At Datnexa, we see common standards and cross-sector collaboration as essential infrastructure for public-sector AI. They can reduce unnecessary duplication, help teams learn from one another and give residents greater confidence that new technology is being introduced with care.

As Julian Patmore’s contribution underlines, progress will depend not only on what individual organisations build, but on what the sector is willing to learn, share and assure together.