Impact Factor: 1.7
5-Year Impact Factor: 1.5
CiteScore: 3.1
UN SDG
Turkish Journal of Fisheries and Aquatic Sciences 2026, Vol 26, Num, 4     (Pages: TRJFAS27583)

An Analysis of a ChatGPT Use Case: Can a Large Language Model (LLM) be Used to Plan Artificial Reefs as a Fisheries Management Tool?

Faik Ozan Düzbastılar 1 ,Tevfih Ceyhan 1

1 Ege University, Faculty of Fisheries, Department of Fishing and Processing Technology, 35100 Bornova, İzmir, Türkiye DOI : 10.4194/TRJFAS27583 Viewed : 10 - Downloaded : 9 No studies have examined the appropriateness of artificial intelligence for the planning of artificial reefs used in fisheries management. This study examined ChatGPT's capabilities in planning artificial reef (AR) projects by asking 50 questions and evaluating the answers from five experts. This approach aimed to assess the interactivity of ChatGPT, its contribution to the advancement of marine science and technology, and its potential limitations in the applied marine context. We analysed the experts` ratings using the Likert scale. Specifically, the appropriateness of responses varied between prompts, indicating different levels of relevance and appropriateness. Likewise, the validity of the information presented in the responses varied significantly, suggesting differences in the accuracy and reliability of the content provided. Additionally, assessments of the overall quality of responses yielded analogous results, highlighting differences in the completeness and effectiveness of responses. Using the seven-point Likert scale, the average score of the experts for the first ten questions, which are basic aspects of the ARs, was 4.6 for agreement, 4.7 for relevance, and 4.6 for quality of response. For the remaining 40 questions, which were based on specific phases of the AR project, the average scores were 4.6 for agreement, 4.6 for appropriateness, and 4.5 for quality. Our results suggest that while ChatGPT can effectively address fundamental issues related to ARs and provide accessible information on project planning steps to a wide range of stakeholders, including NGO staff, ministry engineers, private sector officials, and students, it is less reliable for nuanced, high-level scientific inquiries. In summary, while ChatGPT shows promise as an educational and planning aid in the context of ARs, its application should be undertaken cautiously to mitigate the risks associated with its current limitations. Advances in AI and specialized data access are expected to expand their role in research and project planning, improving utility and reliability. Keywords : Artificial reef Project planning Artificial intelligence Large language model ChatGPT