The integration of artificial intelligence into public administration is shifting from a technical novelty to a structural reality across European governance, raising significant questions regarding accountability and the transparency of decision-making. From local municipal offices in Acqui Terme to diplomatic and administrative discussions in Tirana, government officials are increasingly utilizing generative AI tools to draft policy documents, manage correspondence, and analyze complex datasets, according to recent reports from the European Parliament’s legislative analysis on digital transformation.
While proponents argue that these tools streamline bureaucratic processes and reduce administrative backlogs, critics point to the “black box” nature of algorithmic outputs. In Italy, the deployment of automated systems for public services has triggered a broader debate regarding the EU AI Act, which mandates strict transparency requirements for high-risk AI systems used by public authorities. The core issue remains: when a minister or local assessor relies on an automated draft, the line between human responsibility and machine generation blurs, potentially complicating legal liability for administrative errors.
The Shift Toward Algorithmic Governance
Public administrations are adopting AI for a variety of tasks, ranging from summarizing public feedback to drafting legislative proposals. This transition is framed by many local administrators as a necessary step toward modernization. However, the reliance on AI-generated content has prompted concerns among civil society groups about the potential for algorithmic bias and the erosion of specialized human oversight. According to the OECD AI Principles, public sector entities must ensure that human intervention remains at the center of critical decision-making, particularly when those decisions impact the fundamental rights of citizens.
In practice, the use of AI in local government often lacks standardized oversight protocols. While the European Union has established a comprehensive framework under the AI Act—which entered into force in August 2024—the practical application at the municipal level remains inconsistent. Local administrators are often left to navigate these tools without specific regional guidelines, leading to a fragmented approach where the level of AI integration depends heavily on individual municipal policies rather than a unified national mandate.
Transparency and the Accountability Gap
A primary concern for legal experts is the “accountability gap” created when AI systems fail or produce inaccurate information. If a public official signs a document drafted by an AI that contains erroneous data or biased conclusions, the chain of responsibility is not always clear. Under existing administrative law, the human official remains the primary signatory and is therefore legally liable for the contents of the document. Yet, as systems become more complex, the ability for an official to verify every line of an AI-generated output diminishes.
The Italian Data Protection Authority (Garante per la protezione dei dati personali) has frequently emphasized that the use of automated decision-making processes must be accompanied by “meaningful human involvement.” This ensures that the final authority rests with an individual capable of evaluating the context, ethics, and legal implications of the decision. Without this, there is a risk that administrative acts could be challenged in court on the grounds of lack of transparency or procedural fairness.
Future Outlook and Regulatory Compliance
As the adoption of AI continues to accelerate, the focus is shifting toward how governments will implement internal audit mechanisms to track AI usage. The European Commission has highlighted that public sector entities will need to prioritize training for staff to ensure they can critically evaluate AI outputs. Furthermore, the development of “explainable AI” (XAI) tools is becoming a priority for government contractors, as these systems allow administrators to trace the logic behind a particular recommendation or draft.
The next major checkpoint for these developments will be the full implementation of the EU AI Act’s provisions regarding public sector accountability, which are set to be phased in through 2026. Local and national governments are expected to publish transparency registers detailing which AI tools are being used and for what specific administrative tasks. For citizens, this suggests a future where the influence of artificial intelligence on public life will be more visible, even if the underlying technology remains complex.
What are your thoughts on the use of AI in local government? We invite you to share your perspective in the comments section below or join the conversation on our social media platforms to discuss how we can balance innovation with administrative accountability.
Worth a look