From NHS trusts to central government departments, housing associations to universities — AI adoption in public services must be demonstrably responsible, not just effective. That requires a different starting point from commercial adoption.
The public services sector spans genuine breadth. NHS trusts and integrated care boards managing patient data and clinical workflows. Central government departments including HMRC, DVLA, and the civil service proper. Police forces and fire services dealing with sensitive operational data. Housing associations managing vulnerable populations. Universities navigating institutional AI adoption while simultaneously governing how students and academics use it.
What unites these organisations is not their function but their shared constraints. Public accountability is not a design choice — it's a legal requirement. When a citizen is affected by a public sector decision, they have a right to understand it. GDPR in a public context is stricter than in commercial settings. Union considerations shape how technology can be deployed. Procurement frameworks constrain tool choice. Data sensitivity is high because the stakes for individuals are high.
Against those constraints sits genuine pressure to act. Central government is actively pushing AI adoption for productivity. The NHS is under intense workforce and funding pressure and sees AI as essential to managing demand. Housing associations face mounting expectations around operational efficiency. Universities are adopting AI rapidly whilst wrestling with integrity and assessment questions that don't have simple answers.
The sector is excited about AI but not yet organised to deploy it responsibly at scale.
Central government is pushing hard. The Cabinet Office and Government Digital Service have committed to AI adoption across departments. There's money and momentum. But there's also a real question: which departments have the governance frameworks in place to use that investment well?
The NHS is under intense pressure. Waiting lists, staffing shortages, budget constraints. AI is being positioned as part of the answer — clinical triage, administrative support, scheduling optimisation. But NHS organisations are acutely aware that a single AI failure in a clinical context can be catastrophic. The governance question isn't abstract here. It's a live operational concern.
Housing associations have a clearer efficiency opportunity. Back-office work, tenant communication, case management. But they're often resource-constrained — a housing association doesn't have the technology capacity of a major commercial company. Governance frameworks, though necessary, are real friction when capacity is limited.
Universities sit in a unique position. They're adopting AI for institutional operations — research support, student services, administrative efficiency — whilst simultaneously having to govern how students and academics use AI tools. These are fundamentally different governance problems requiring different approaches.
The common thread is honest: organisations are enthusiastic about AI but haven't yet resolved the governance questions that would let them deploy with confidence at scale. There's no shortage of tools or ambition. There's a real shortage of clear governance frameworks, risk assessment methodologies, and structured approaches to responsible deployment.
Every public organisation has different governance requirements, different political contexts, and different service pressures. The engagement always starts with understanding yours.
Whether you're an NHS trust exploring clinical workflow AI, a government department building its governance framework, or a housing association looking at operational efficiency — the approach is structured around your accountability requirements, not a generic playbook.
A governance-informed 2–4 week assessment that maps where AI can create genuine value, identifies the specific governance questions that need resolving before deployment, and recommends realistic pilots. This isn't a consultant's shopping list. It's discovery work that respects the constraints of public sector procurement, data sensitivity, and accountability requirements. You emerge with a clear opportunity map, a risk and governance assessment, and a pilot recommendation that can withstand scrutiny.
A 6–12 week structured pilot in a real operational context. Not a sandbox. Not a proof of concept. An actual operational pilot where governance is built in from day one. For public organisations, this often means partnering with one directorate or service, embedding an operational team, and building the audit trail and risk controls that let you scale. The pilot generates both usage data and governance evidence — both essential for public sector stakeholder buy-in.
A 3–6 month engagement to move from successful pilot into sustainable operations. For large public organisations — NHS trusts, government departments, housing associations with multiple locations — this is where the real complexity lives. Clear governance frameworks, training pipelines, role clarity, escalation routes, and monitoring. You need to be able to explain to elected representatives, union representatives, and regulators why this AI system is being used and how it's being controlled. We embed with your team to build the operating model that makes that possible.
A retained engagement — typically 1–3 days per month — providing senior AI oversight, board-level governance advice, and stakeholder credibility. For many public organisations, the challenge isn't implementation. It's having a trusted figure with sector experience who can navigate governance frameworks, manage stakeholder relationships, and help the board understand the genuine risks and opportunities. Mike can serve as fractional head of AI or chief AI adviser, often without requiring a permanent hire.
Mike was part of the senior leadership team for Cisco's £430M EMEA Public Sector business, leading the Local Government and Education vertical. That meant working directly with councils, education authorities, and the procurement frameworks that shape technology decisions across the sector.
This isn't general familiarity with the public sector. It's direct experience of how large public organisations evaluate technology, navigate procurement, build business cases, and manage the governance that surrounds significant technology investment. The disciplines of public sector selling at that scale — accountability, transparency, value for money, stakeholder management — translate directly into how AI adoption needs to be approached.
Not an advisory engagement. An elected practitioner role with democratic accountability for AI adoption inside a local authority. The work is directly relevant to any public sector organisation navigating the same governance challenges:
Responsible AI Policy & Strategy: Co-authored both the council's formal Responsible AI Policy and its AI Strategy — establishing governance frameworks, ethical principles, an AI Steering Group, approval processes for new use cases, and clear escalation routes. Both ratified at Cabinet level.
Microsoft Copilot rollout: Oversaw deployment to approximately 250 officers and elected members — one of the earlier council-wide AI tool deployments in English local government.
Operational AI pilots across services: AI assistance deployed into planning (application processing, consultation analysis, property retrofit assessment), democratic services (committee minutes, agenda preparation), and customer services (resident self-service, communications drafting). Real operational use in a governance-constrained environment, not sandbox experiments.
This is the governance experience that translates across public services — the same questions of accountability, transparency, union relations, equalities impact, and public trust apply whether the organisation is an NHS trust, a government department, or a housing association. See the full local government detail →
The governance frameworks Mike built at Verimatrix — delivering measurable impact including ~$1M annualised engineering productivity gains and ~$3M revenue uplift — were designed for a regulated, multi-jurisdiction environment. The principles transfer directly into public sector contexts where regulatory compliance and board accountability are equally non-negotiable: clear risk assessment, defined decision-making, audit trails, and regular review.
Governance-first, operationally grounded AI adoption for public services.