随着海洋大数据时代的来临和人工智能技术的蓬勃发展，人工智能海洋学作为新兴的交叉学科逐渐崭露头角。该学科深度融合海洋科学和人工智能技术，为海洋研究和应用提供了新的方向和强有力的科技手段。然而，在人工智能海洋学发展初期，我们也面临一系列挑战。

为了应对当前的诸多挑战，本研究系统梳理人工智能海洋学在建立海洋大数据系统、开发适用于海洋大数据的人工智能算法、海洋智能大模型和海洋数字孪生系统等方面的最新进展。研究发现，人工智能海洋学在数据整合、存储与分析领域正迅速发展，不仅拥有庞大的数据资源，还积累了丰富的数据处理经验。借助深度学习、机器学习等前沿技术，人工智能算法能够更加智能和高效地解读和应用海洋数据，为海洋科学研究开辟了新的可能性，展现了广阔的应用前景。

人工智能海洋学已涉足多个关键领域，包括海洋特征智能识别、海洋要素与现象智能预报、模式参数智能估算等。智能识别海洋特征使我们更准确地了解海洋环境的动态变化，为海洋环境安全提供更为精准的保障。基于人工智能算法的预报模型提高了海洋气象、海洋生态和动力过程等方面的预报精度，为海洋生态环境保护和资源可持续利用提供了可靠的支撑。同时，智能估算海洋模式中的各类参数，可节省计算资源并提升模式的模拟和预报能力。

展望未来，我们需要研发定制化的自然语言与信息处理技术(如Ocean ChatGPT), 需要发展基于人工智能的地球系统耦合大模型，需要进一步夯实海洋数字孪生系统的技术框架和数据底座，需要在算力调度、硬件资源、计算效率和稳定性等方面下功夫，为人工智能海洋学这一海洋科学的革命性分支打下坚实的基础。随着人工智能技术的迅猛发展和相关领域人才队伍的不断壮大，人工智能海洋学的发展前景将远超我们的想象，将为人类探索海洋的奥秘和应对全球气候变化等重大问题提供更为强大的、智慧高效的工具。

With the advent of the era of ocean big data and the flourishing development of artificial intelligence (AI) technology, AI oceanography has emerged as a new interdisciplinary science. By deeply integrating marine sciences with AI technology, it can provide strong support to the field of oceanography in a totally unprecedented manner. However, AI oceanography is still in its infancy and is facing many challenges, such as the lack of clarity in the direction it goes, in the scientific problems it addresses, and in the planning of its development pathways. 

This study reviews the current state of AI oceanography, including new advancements in the establishment of ocean big data systems, the development of relevant AI algorithms, and the construction of the digital twins of the ocean. It is found that AI oceanography is rapidly advancing in the areas of data integration, storage and analysis, accumulating not only vast data resources but also rich experience in data processing. Via cutting-edge technologies such as deep learning and machine learning, AI oceanography enables more intelligent analysis and application of ocean data, proving new research opportunities and possibilities. 

In terms of applications, AI oceanography has made significant strides in research areas such as intelligent recognition of ocean features, intelligent prediction of ocean phenomena, and intelligent estimation of model parameters. Intelligent recognition of ocean features allows for a more accurate understanding of dynamic processes, providing precise safeguards for marine environment. AI-based intelligent models can enhance the prediction accuracy for ocean meteorology and marine ecosystems, offering reliable support for marine environment protection and sustainable resource utilization. Intelligent estimation of model parameters can largely increase models’ computational efficiency as well as their simulation and forecast ability.

Looking ahead, we need to develop customized natural language processing technologies (i.e. Ocean ChatGPT), to build AI-based Earth system big models, to further solidate the framework and database for the digital twins of the ocean, and to work collectively on computing power scheduling, hardware construction, computing efficiency and stability, providing a robust foundation for AI oceanography, a truly revolutionary branch of ocean sciences. With the rapid development of AI technology and the continued growth of research teams, AI oceanography will go far beyond our imagination, providing powerful intelligent support for us to explore the mysteries of the ocean and address critical issues such as global climate change.