Name in original language
MI integrācija akreditācijas procesā
Initiative overview
This initiative addresses the growing complexity and workload of accreditation activities carried out by the Latvian National Accreditation Bureau (LATAK). Accreditation requires the review and preparation of large volumes of documents while complying with national, European and international requirements. Over time, this has led to a heavy reliance on manual work by highly qualified staff, creating risks to efficiency, consistency and the ability to meet strategic objectives.
The initiative leverages artificial intelligence as an opportunity to support staff in routine and time-consuming tasks, allowing them to focus more on expert judgement, quality assurance and strategic work. The main objective of the initiative is to improve the efficiency, quality and sustainability of accreditation processes while maintaining full human oversight and accountability.
AI tools are used to support activities such as drafting and reviewing documents, preparing correspondence, summarising information and supporting meeting preparation. By reducing time spent on repetitive tasks, the initiative aims to improve turnaround times, enhance the clarity and consistency of outputs, and reduce the risk of delays or quality gaps. At the same time, it supports staff development by strengthening digital skills and fostering a more open and innovative organisational culture.
The initiative has been implemented gradually through controlled testing, pilot use cases and practical workshops. A key principle is the use of officially approved AI tools within a secure infrastructure, combined with clear rules on data protection and mandatory human review of all AI-supported outputs. Staff engagement and training are central elements, ensuring that AI is understood as a support tool rather than a replacement for professional expertise. This approach helps build trust, manage risks and embed AI use into everyday work responsibly.
Looking ahead, the initiative is expected to evolve from individual productivity support towards more integrated and institutionalised solutions. Planned next steps include exploring additional AI tools, developing low-code or no-code digital platforms to support assessment and decision-making, and strengthening risk-management and governance frameworks. The approach is designed to be scalable and transferable, allowing similar public institutions to adapt the model to their own processes while maintaining transparency, security and human-centred AI use.
