Introduction: To predict the future, how a coming day, week, or month will look has become even more crucial than ever for a firm, due to recent pandemic crises, and wars. Being able to predict the future will enable firms to reduce costs and increase time efficiency. Processes such as forecasting have been at the forefront to aid managers in these matters by improving decision-making and planning. Greater forecasting capabilities have been achieved by adopting technologies such as Artificial Intelligence (AI). As it has shown to aid practitioners in predicting the future with high accuracy. Thus, leading to improved decision making and planning. Problem discussion: AI is still in its infancy, and technology adoption is a staged-based process. More research is needed to identify the potential barriers a firm faces when looking to adopt AI into their forecasting process. As well as how these barriers are mitigated, and what barriers are relevant depending on the stage of adoption. Purpose and Research question: The purpose of this study is to investigate the barriers of AI pre-adoption in forecasting and how these barriers are mitigated. To answer the following research question: How does a manufacturing firm mitigate AI pre-adoption barriers in the forecasting process? Method: First, a scoping review is conducted to identify barriers in AI adoption with the support of the TOE framework, (Technological, Organizational, and Environmental). Later, the thesis follows a qualitative approach, conducting a single case study. The primary source of empirical data was collected from five in-depth semi-structured interviews. The data is collected from an international manufacturing firm located in Sweden that is looking to adopt AI-ML into its forecasting process. The findings collected from the firm are later discussed with an expert in the field of AI and forecasting to further bring validity and input to the findings. Findings: Organizational readiness, Top management, Poor data, Inappropriate technology infrastructure, and Partnership were identified as key barriers in the AI-ML pre-adoption for the forecasting process. The barrier could be mitigated by building a strong business case, creating managerial awareness and understanding, interactive data platform, comprehensive dataset, and incentives. Conclusion: The study provides theoretical contributions as well as managerial implications. By shedding light on the barriers in the pre-adoption phase and providing insight as to how to mitigate the barriers. Future research is recommended to study the same phenomena at another firm.