Generative AI Applied to Technology Strategy - Using Open-source Large Language Models In an Agentic Way

Abstract

This thesis explores the application of generative AI in technology strategy, specifically through the use of open-source Large Language Models (LLMs) to develop an agent capable of understanding documents and providing insightful analysis. The research focuses on adapting existing LLMs to automate the extraction, synthesis, and analysis of external data from diverse sources. By implementing techniques such as Retrieval-Augmented Generation (RAG), we aim to enhance the accuracy and reliability of AI-generated content, mitigating issues like hallucination. The project involves the customization of LLMs for strategic decision-making, the development of a user-friendly interface, and rigorous evaluation to ensure operational effectiveness. Our findings indicate that integrating generative AI with strategic planning processes can significantly improve decision-making efficiency and accuracy, offering valuable insights for the electrification of road transportation and beyond.