Artificial Intelligence (AI) is transforming how the public sector operates. Trials are already underway across public sector departments, including within the NHS, following a recent National Commission which will accelerate access to AI assistants for doctors.
Momentum is building quickly and is supported by new government-offered guidance on using AI safely in the civil service through its AI Playbook. Following this, the UK Government announced plans to integrate AI software into core departments across the board, having signed a partnership with Open AI in July. However, as the uptake of AI in the public sector increases, a critical issue is being overlooked: data integrity. Alarmingly, 45 percent of public sector organisations lack a formal data strategy altogether, and risk fuelling their AI models with untrustworthy inputs.
A common misconception is that AI software is impartial. However, AI tools rely entirely on the trustworthiness of the data they are fuelled by. The UK public sector therefore urgently requires a comprehensive strategy, led by a chief data officer (CDO) going into 2026. The CDO should oversee the collection, governance, and use of data to ensure its quality, and extract its value throughout the organisation. Without this foundation, the public sector will not truly unlock the benefits of AI.
An urgent issue: the lack of data integrity in the public sector
In the public sector, the consequences of faulty or biased AI are potentially life-altering, leaving no room for error. We have seen real-world impacts of this over the past year, raising serious concern and consequential damages to trust in the adoption of AI in public sector bodies.
For example, recent findings indicate that AI tools like Google’s Gemma, used by over half of England’s councils, are diminishing women’s health issues and ultimately risk creating gender bias in care decisions. The production of incorrect and biased outputs due to incomplete, unreliable, and inaccurate data is already having serious implications within the healthcare sector. Without a CDO to regulate the data flowing into, out of and throughout an organisation, these biases will continue to emerge.
In contrast, 71 percent of organisations with governance programmes report high trust in their data, compared to 50 percent without. Ensuring sound data quality directly boosts organisational confidence in decision-making when using AI and – given the millions of citizens served by the public sector – confidence in its decision-making processes is crucial.
Evolving AI and emerging AI agents demand a CDO in 2026
By establishing leadership over a strategy, a CDO can effectively unlock data-driven initiatives, including generative AI, that are central to the UK public sector’s efforts to drive transformation in the coming year. For example, recent findings have shown that five government departments are using AI to draft responses to questions in Parliament. The demand for high-integrity data has never been greater, as these tools are handling vast amounts of sensitive government information, and personal citizen information.
The imminence of agentic AI’s implementation within government practices makes this issue increasingly urgent. Unlike traditional generative AI systems and LLMs – which are trained on text and language data to generate text-based outputs in response to prompts – these agents will act with autonomy, within the parameters controlled by the user. These implementations will soon be felt, as, by 2029, agentic AI software is expected to autonomously resolve 80 percent of common customer service issues without human intervention.
As more agents are introduced, autonomy combined with guardrails, transparency, and flexibility is crucial to give the public sector the ability to innovate, while ensuring their data integrity processes are working continuously in the background. The CDO and their implemented strategy will ensure human oversight – and accurate, consistent, and contextual data to fuel AI tools.
By guaranteeing a data-driven culture, and applying robust governance and quality standards, a proactive CDO mitigates risk. Public sector bodies can then be confident in these AI-driven insights to aid critical decision-making, enhance operational performance and drive economic growth.
Driving AI transformation with data integrity
To further ensure meaningful AI outputs, the public sector must also reevaluate how it stores data. Many organisations within the public sector rely on multiple applications to host different types of personal and sensitive data, storing them in varying internal systems. For example, a recent UK government report found that 21 of the 72 highest–risk legacy digital systems prioritised in the 2022–2025 digital and data roadmap still lack funding. Data quality and data sharing barriers are persistent and long–standing.
This is an issue that requires urgent attention through foundational elements of a data integrity strategy, which is built on four pillars:
- Enterprise-wide integration
- Location intelligence
- Data enrichment
- Accuracy, governance and quality
To address the current issue and ensure meaningful AI outputs, the public sector must firstly integrate data across cloud, on-premises, and hybrid environments, as well as across different departments. At present, data is often inaccessible and locked within outdated legacy IT systems. Building a holistic view of the data provides AI models with a more comprehensive understanding of trends, leading to well-informed and accurate results.
Secondly, context is key. Without context, even high-quality data risks misinterpretation. Public sector organisations must ensure they enrich first-party data with curated third-party sources – including demographic profiles, precise address data, and environmental risk indicators. This external context allows for valuable insights to emerge, enabling more informed decisions.
Finally, departments require a solution that automates governance and stewardship tasks, ensuring the quality, value, and trustworthiness of their data. Trusted high-quality datasets are essential to powering AI systems that deliver accurate outcomes. This is particularly important in public services, where providing the correct information to the right citizens is critical as errors could lead to GDPR violations.
By implementing these four pillars of data integrity, the CDO can ensure that AI use is supported by a trustworthy data foundation and empower public sector departments to act with confidence in their data.
The future of data integrity in the public sector’s adoption of AI
The growing use and advancements of AI in public sector bodies is driving urgency to establish sound data integrity strategies. In 2026, public sector organisations must appoint more strategic roles, including a CDO, to ensure trusted data flows throughout, and emerging AI technologies are implemented successfully and ethically.
Data integrity is essential for generating meaningful insights, on which important decisions can be made in the UK public sector. Without this, the full potential of AI cannot be realised, and organisations risk causing harm and perpetuating bias within UK society.
