Introduction
The integration of artificial intelligence and emerging technologies into fisheries science has fundamentally transformed marine resource management approaches (Bradley et al., 2019; Ebrahimi et al., 2021). This field has evolved from foundational object-oriented modeling approaches (Bousquet et al., 1994) to sophisticated expert systems such as CANOFISH and ProTuna, which have enhanced management decision accuracy by 85% (Alagappan and Kumaran, 2013). Currently, the integration of emerging technologies including satellite image processing, smart sensor networks (WSN), and deep learning algorithms has created a new paradigm in sustainable marine resource management (Lu et al., 2024). Global fisheries face significant challenges, as FAO reports indicate 94% of aquatic resources are in two distinct states: 60% in full exploitation and 34% at levels beyond biological sustainability (Kumar et al., 2024; Stroe, 2024). This situation is directly related to illegal, unreported, and unregulated fishing (IUU), which accounts for 20-35% of global catch and causes annual economic damages of $10-23.5 billion (Samy-Kamal, 2022; Grey, 2023; Lubchenco and Haugan, 2023). Fishing activities impact not only target species but also non-target species and biodiversity (Liang and Pauly, 2017), while socioeconomic factors contribute additional complexity to fisheries management (Phillipson and Symes, 2013). This review examines technological advancements in fisheries management from 2004-2024, focusing on machine learning developments in conjunction with traditional management approaches. The investigation addresses how artificial intelligence has improved management efficiency, what implementation challenges exist across different contexts, and what frameworks are necessary for sustainable integration of AI in global fisheries management.

Methodology
 This study applies a systematic review methodology that comprises both quantitative and qualitative methods to examine the implementation and effectiveness of AI technologies in fisheries management. The research procedure was a three-phase structured method that started with a full-fledged literature search in the most important scientific databases including the Web of Science, Scopus, and Google Scholar, with a time frame of 2004 to 2024. Then, the investigation proceeded with the analysis of technical reports from international organizations such as FAO and the World Bank to gain an understanding of the practical aspects of the project, as well as broad analyses of case studies from both developed and developing countries to observe real-world implementations and problems. Data analysis included statistical evaluation of the implementation results via comparisons of success rates over different areas and thematic analysis of the implementation challenges. The main point is, studies brought about objective evaluation of the technology impact among the different locations.
Results
Artificial intelligence and emerging technologies have demonstrated significant contributions to fisheries management. In monitoring applications, empirical studies show that machine learning applied to fish species identification from images has achieved 95% accuracy (Silva et al., 2022). Additionally, the integration of Automatic Identification System (AIS) and Vessel Monitoring System (VMS) data has led to a 40% improvement in marine spatial planning (Thoya et al., 2021; Lu et al., 2024). Recent research demonstrates that deep learning models in early detection of environmental threats have accuracy above 90% (Fei et al., 2023), while advanced radar technologies in monitoring wildlife interactions and fishing activities have shown remarkable efficiency (Navarro-Herrero, 2024). The scalability of these solutions has been enhanced through the development of open-source frameworks, enabling traditional fisheries to benefit from advanced technologies (Silva et al., 2022). Implementation challenges span technical domains, with data standardization issues prominent; socioeconomic barriers, which vary significantly between regions; and regulatory constraints, characterized by adaptation delays.
 Discussion and conclusion
The transformative potential of artificial intelligence in fisheries management requires balanced consideration of technical, socioeconomic, and institutional factors for successful implementation. Studies have shown that the integration of remote sensing data with AIS can effectively monitor IUU fishing activities, particularly in regions with limited monitoring capacity (Kurekin et al., 2019). Smart technology implementation in aquaculture has led to significant efficiency improvements through IoT systems and smart sensors, demonstrating the economic value of AI integration (Lan et al., 2022). Local ecological knowledge (LEK) complements scientific data by providing deeper understanding of marine ecosystems (Silvano and Valbo‐Jørgensen, 2008). Successful examples include identification of causes for fish population decline (Dey et al., 2019) and bycatch management (Cazé et al., 2022). The scalable framework for fish image collection and annotation proposed by Silva et al. (2022) demonstrates how technology can be made accessible across different contexts. Three principal directions for future development are identified: standardization of integration protocols, capacity development in developing regions, and adaptive regulatory frameworks. Future initiatives should address implementation barriers, develop comprehensive training programs, and establish regulatory frameworks that facilitate innovation while ensuring sustainable resource management.
Conflict of Interest
The authors declare that there is no conflict of interest in this research work.
Acknowledgment
We sincerely thank the Office of Vice Chancellor for Research and Artemia and Aquaculture Research Institute of Urmia University for the kind support.