Report for the Fifth Machine Learning in Geotechnics Dialogue (5MLIGD) on “Challenges and Experience in Geo-Data Sharing”

This report summarizes key outcomes of the Fifth Machine Learning in Geotechnics Dialogue (5MLIGD), themed “Challenges and Experience in Geo-Data Sharing.” Recognizing the critical role of data-sharing practices in geotechnical engineering, invited experts explored emerging technological frameworks, including Web3.0 blockchain infrastructures and differential privacy techniques, as potential solutions for enhancing data governance, privacy, and trust. Group discussions and questionnaire analyses revealed widespread consensus on the benefits of geo-data sharing, notably improved risk assessment, innovation, and informed decision-making. However, substantial concerns remain regarding data security, privacy breaches, and misaligned incentives between academia and industry. Recommendations derived from the dialogue emphasize a multi-faceted approach: adopting advanced privacy-preserving technologies, developing standardized data formats, creating secure, user-friendly platforms, and establishing incentive-based policies. Additionally, targeted education and the cultivation of a collaborative community culture are critical to overcoming barriers. Collectively, these initiatives aim to transform geo-data sharing into a secure, mutually beneficial norm, ultimately fostering innovation, efficiency, and resilience in geotechnical engineering practice.