DREAM AI Assistant for public investment management

The problem

Ukraine's reconstruction amid an ongoing war is one of the most complex public investment challenges in modern history. International estimates put the cost at over 588 billion dollars. Thousands of projects. Communities, regions, ministries, and dozens of international partners — all working simultaneously.

Ukraine’s public investment management reform has created unified digital workflows for project preparation, appraisal, prioritisation, funding coordination and monitoring. DREAM is the digital backbone of that reform, supporting more than 12,900 public investment projects across 18 sectors and used by more than 1,300 municipalities, all regional administrations, all ministries and major state-owned enterprises.

That scale brings a practical capacity challenge. Public officials must navigate complex laws, methodologies, procurement rules, construction norms, financial requirements and donor conditions, often with limited staff and under wartime pressure. For example, a full cycle of investment pre-feasibility study takes about 160 hours, or four weeks of full-time work for a single local government employee. As a result, preparing 10 projects becomes a huge challenge for a small local government. During peak planning cycles, the DREAM Project Office handled more than 850 support requests per day, even with extensive training, helpdesks and expert networks in place.

The solution

The DREAM AI Assistant is being developed as a support layer inside the DREAM ecosystem with a staged rollout from 2026 to 2028. In the first phase, it provides real-time, context-specific guidance on public investment management rules and project preparation directly within the system interface. The aim is to translate complex legal and methodological requirements into practical explanations tailored to a user’s role, sector and location.

Later phases are expected to support quality assurance by flagging missing indicators or inconsistencies in feasibility studies, comparing submissions with historical cases and, eventually, supporting scenario-based investment modelling.

The core of the system utilises Retrieval-augmented generation (RAG), which ensures that the AI's responses are grounded exclusively in verified laws, manuals and guidelines. To safeguard sensitive information, the system employs local language models hosted on secure infrastructure, maintaining a high standard of data sovereignty. Furthermore, the architecture includes a human-in-the-loop validation process to ensure that all content remains accurate and reflects the most recent regulatory updates.

To minimise the risk of AI hallucinations or the use of outdated guidance, every response generated by the system is traceable to an approved source and marked with a specific timestamp and version. Potential risks, such as factual errors or privacy concerns, are proactively addressed through structured prompt design and rigorous access controls.

A dedicated multidisciplinary oversight team provides continuous monitoring to ensure the system adheres to ethical standards and remains focused on its public purpose.

The impact

DREAM is used by 1360 local self-government bodies — 90% of all Ukrainian communities, all 24 regional administrations, all line ministries - to plan and manage their projects and provide communities a direct channel to rebuild according to their needs. Ukraine’s Single Project pipeline with more than 17,000 projects, is open and publicly visible at dream.gov.ua.

Lessons learned

• Clear rules, workflows, verified localised knowledge and careful design are critical. AI is helping determine how knowledge (policies, methodologies, requirements, sectoral expertise, etc.) is accessed, interpreted, and applied – especially in a fast-evolving, complex reform environment.
• Organisational readiness and a responsible AI use framework are important. The approach is grounded in a working prototype, a curated and verified knowledge base, and a hybrid architecture aligned with Ukraine’s data sovereignty and ethical AI principles stated in White Paper on Artificial Intelligence Regulation in Ukraine.
• In high-stakes public investment, the human in the loop is critical. A dedicated multidisciplinary oversight team provides continuous monitoring to ensure the system adheres to ethical standards and remains focused on its public purpose.

Ukraine’s public investment management reform includes amendments to the Budget Code, strategic planning rules, investment project preparation and appraisal procedures, procurement rules, construction norms, financial and fiscal requirements, and donor conditions. The AI Assistant is also being designed in line with Ukraine’s data sovereignty and ethical AI principles.