As artificial intelligence (AI) becomes increasingly embedded in public sector operations, a critical question emerges: should AI be treated as a tool to automate tasks, or as an autonomous team member capable of agency and collaboration?

1. AI as an Automation Tool

In many government departments, AI is already streamlining operations, automating repetitive tasks, analysing large datasets, and improving service delivery. These systems are designed to be predictable, rule-bound, and efficient. They act like diligent clerks: fast, tireless, and precise.

This model is particularly effective in stable environments where compliance, consistency, and transparency are paramount. AI tools can reduce human error, free up staff for strategic work, and improve turnaround times for citizen services. For example, robotic process automation (RPA) paired with AI can handle claims processing, document classification, or fraud detection with minimal human intervention.

2. AI as an Autonomous Team Member

But what happens when AI begins to act with agency?

Emerging research suggests that AI can do more than automate, it can collaborate. Autonomous agents are now capable of making decisions, adapting to changing conditions, and even facilitating human teamwork. These systems resemble senior consultants: they learn from experience, reconfigure workflows, and offer alternative perspectives.

In public service contexts, this could mean AI helping to shape policy recommendations, manage stakeholder engagement, or coordinate multi-agency responses. The COHUMAIN framework, developed by Carnegie Mellon University, explores how AI can strengthen human collaboration rather than replace it, nudging team members, surfacing blind spots, and enhancing interpersonal dynamics.

Ethical and Operational Implications

Treating AI as a team member introduces complex challenges. Who is accountable for its decisions? How do we ensure transparency, fairness, and inclusivity? Public servants must grapple with new governance models, ethical frameworks, and training needs.

Government playbooks increasingly stress responsible AI deployment, emphasising oversight, explainability, and citizen trust. The UK’s AI strategy, for instance, calls for human-centred design and robust safeguards to ensure AI supports, rather than undermines, democratic values.

A Hybrid Future

Rather than choosing one model over the other, many experts advocate for a hybrid approach. AI can be both a powerful tool and a collaborative partner, depending on the task, context, and organisational goals.

For public servants, the key is to design systems that respect human agency while leveraging machine intelligence. This means investing in digital literacy, ethical leadership, and inclusive design. It also means asking not just what AI can do, but what it should and shouldn't do.

Suggested Reading:

UK Government AI Playbook

COHUMAIN Framework by Carnegie Mellon

Leaders’ Guide to AI for All

All points mentioned here are my personal views, based on interest, intellectual curiosity and current studies

Make sure to share your own thoughts with the author by leaving a comment below