In spite of the efforts to become more data-driven, organizations still need to overcome common data governance challenges such as interoperability, data sovereignty, and data value generation. On top of this, the large volumes of data being generated in the computing continuum generate innovative business models. In this regard, data spaces facilitate the safe exchange of data assets to improve decision-making, foster innovation, and create novel services, products, and business models. The standardization, integration, cleaning, and transformation of the various data sources is crucial for delivering reliable data assets. To this end, data harmonization is key as it raises data quality and usability, reduces redundancy, and helps organizations meet regulatory and industry standards. In this manuscript, we dive into the scientific literature to better understand the various stages that comprise the data harmonization lifecycle and the challenges of it in the field of data spaces. Then, we analyze the various artificial intelligence techniques utilized for data harmonization and its role as a standardizing agent for data definition and interoperability, looking at the current studies and the overwhelming related regulation.