Imagine effortlessly accessing complex land data using simple, everyday language. That's the promise of new research exploring how AI can transform how we interact with databases like the Land Matrix, a vital resource for understanding land acquisitions worldwide. Researchers are tackling the challenge of making this valuable data more accessible to policymakers and the public, who often lack the technical expertise to navigate complex database queries. The key lies in adapting Large Language Models (LLMs), the technology behind AI assistants like ChatGPT, to understand natural language questions and translate them into the precise queries needed to extract relevant information. This isn't as simple as it sounds. LLMs face hurdles in grasping the nuances of human language and the intricacies of database structures. The research compares various leading LLMs, including Llama 3, Mixtral, and Codestral, and evaluates different optimization techniques like prompt engineering, retrieval augmented generation (RAG), and AI agents. These methods aim to enhance the LLMs' ability to understand the context of the questions and generate accurate queries. The results reveal that Codestral, combined with a multi-agent approach, performs best, successfully translating a significant portion of user queries into actionable database requests. While challenges remain, this research demonstrates the potential of AI to democratize access to crucial land data, empowering informed decision-making for a more sustainable future. The research opens exciting possibilities, paving the way for a future where anyone can unlock valuable insights from complex datasets using just plain language.