3.1 Introduction
Artificial intelligence (AI) is a branch of computer science focused on creating intelligent systems capable of performing tasks typically requiring human intelligence, such as reasoning, learning, problem-solving, visual perception, and language processing (Glover Reference Glover2024). Unlike traditional data-processing methods, which tend to be less analysis-oriented, AI systems learn from data. The more data AI systems collect, the better they learn and perform tasks as humans do (Obschonka and Audretsch Reference Obschonka and Audretsch2020). AI is recognized as an intelligent system capable of processing vast quantities of data known as big data and extracting relevant information from it. Big data and AI complement each other. This involves extensive sources of both structured and unstructured data from various origins, including sensors, social media, and transactions, and making sense of them. AI technologies employ machine learning, neural networks, and deep learning to identify patterns within big data, predict outcomes, and generate new knowledge. AI can undertake complex tasks that would otherwise require human effort, such as making decisions in uncertain situations, learning from experience to improve performance, and generating new data, as seen with generative AI models (Liu Reference Liu2021).
The maritime industry, which is the backbone of global trade and the economy, is making tremendous efforts to adopt AI to enhance efficiency and safety and to fuel optimization, predictive maintenance, voyage optimization, and energy management (Durlik et al. Reference Durlik, Miller, Kostecka and Tuński2024). Big data analytics platforms process vast amounts of ship performance and navigation data to support decision-making, ranging from logistics optimization to energy efficiency and safety improvements (Munim et al. Reference Munim, Dushenko, Raaness, Westerberg and Hossain2023). A study by Zhang and Martinez indicates that approximately 35 percent of newly built merchant ships currently utilize AI-powered navigational systems (Zhang and Martinez Reference Zhang and Martinez2023). The most notable applications in this area are automated identification systems supported by machine-learning capabilities, which allow for improved tracking and collision avoidance. These systems utilize inputs from various sources, such as radar, weather forecasts, and satellite imagery, to enhance the predictive accuracy of potential navigational hazards. Nevertheless, they still operate under human oversight rather than as fully autonomous solutions (Johannsen et al. Reference Johannsen, Smith and Rodriguez2024).
This technological advancement is expected to significantly impact the maritime workforce, including seafarers and shore-based personnel. Conventional skills may need to evolve alongside emerging technological competencies, demanding a paradigm shift in maritime education and training. Research conducted by the World Maritime University in 2023 revealed changes in seafarers’ task performance, including the introduction of new roles and responsibilities, modifications to existing activities, adaptation to managing emerging technologies, and the need for diverse new vocational skills and knowledge (WMU 2023). A notable example is the evolution of navigational aids on a ship’s bridge. The study highlighted the transitions occurring across various maritime professions. This transition is not limited to navigation but also extends to engineering and maintenance operations. AI-based technology is increasingly being used to redefine, predict, and optimize ship maintenance, helping to reduce corrective maintenance costs (Gupta Reference Gupta2024). However, this transformation may require engineers to develop new skills in data analysis and the management of systems that generate and process large volumes of data (WMU 2023).
The maritime industry has sought to address various disruptive changes while highlighting the emerging issues facing the maritime workforce. The International Maritime Organization (IMO) published an e-navigation strategy to facilitate the use of digital tools on board, as detailed in MSC 85/26 (IMO 2008). The circular outlines a vision for the future of navigation systems and vessel traffic information, aiming to establish harmonized data and communication frameworks to mitigate the increasing challenges associated with the safe and efficient operation of ships. In this context, the IMO conducted a scoping exercise for Maritime Autonomous Surface Ships (MASS) from 2017 to 2021 to assess the impacts of rising levels of automation on safety, security, and the environment, completing the e-navigation plan.
From the perspective of technology developers, integrating digitalization, information, and communication technologies, including AI, into the maritime industry and onboard ships is a high priority for enhancing safety and operational efficiency. Research reveals that while these advancements offer numerous benefits, they have also introduced various challenges, such as technostress and mental health concerns among seafarers (Lagdami Reference Lagdami, Ahram and Karwowski2024). Moreover, digitalization and ICT depend on reliable connectivity, which increases the risk of cyberattacks. Therefore, such advancements can sometimes be counterproductive and ineffective. As a result, the impact of these emerging technologies on growth and employment is more complex than traditionally perceived and heavily depends on the institutional and policy context (Aghion et al. Reference Aghion, Antonin and Bunel2019).
With the advancement of autonomous ships, scholars hold differing opinions on the future role of seafarers. Some seem to support the idea that automation may reshape rather than reduce employment opportunities (Fonseca et al. Reference Fonseca, Lagdami and Schröder-Hinrichs2021), while others argue that the AI-driven revolution will not necessarily threaten jobs but is likely to lead to a more skilled workforce (Aghion et al. Reference Aghion, Antonin and Bunel2019). This latter perspective is backed by studies such as the one conducted by the Hamburg School of Business Administration (HSBA) for the International Chamber of Shipping, focusing on seafarers and digital disruption. That study indicates that the advent of autonomous ships would redefine rather than eliminate seafarers’ roles, increasing the demand for skilled professionals, particularly marine officers, over the next twenty years. While crew size may adapt due to technological advancements on board, the maritime industry may also see significant additional shore-based job opportunities that require maritime expertise (HSBA 2018).
The preceding discussion highlights the apparent advantages perceived by all regarding the integration of AI in the maritime sector. However, a critical yet often overlooked aspect is the utilization of big data and algorithms associated with AI systems onboard ships, particularly concerning the surveillance and control responsibilities assigned to seafarers. Therefore, this chapter primarily examines the application of AI in the maritime sector and its potential impact on seafarers, given the currently limited use of AI on ships. It also analyzes the legal dimensions and their ramifications for the maritime sector, focusing specifically on the Maritime Labour Convention (MLC)Footnote 1 and other legal instruments regulating maritime labor, such as the International Convention on Standards of Training, Certification, and Watchkeeping for Seafarers (STCW).Footnote 2
3.2 AI Applications in the Maritime Industry and the Implications for Seafarers
The application of AI in maritime operations has accelerated significantly since early 2020. The maritime industry has increasingly adopted AI-powered systems for autonomous shipping, predicting maintenance needs, enhancing safety through real-time monitoring, and conducting risk management, among other functions (Spire Maritime Reference Maritimen.d.). These systems process vast amounts of data from sensors, historical trends, and real-time environmental conditions, integrating them into decision-making processes that were previously reliant solely on human judgment. This shift suggests a future where ships may operate independently without human intervention, introducing challenges to traditional and conventional labor structures and frameworks.
3.2.1 AI and Autonomous Shipping
AI enables the operation of autonomous ships that can operate with few or no crew members on board, utilizing advanced technologies for navigation, safety, and efficiency. As classified by the IMO, MASS are divided into four degrees of autonomy: degree one (automated processes and decision support), degree two (remotely controlled ship with a crew on board), degree three (remotely controlled ship without a crew on board), and degree four (fully autonomous ship; Askari and Hossain Reference Askari and Hossain2022). AI is essential for MASS operations, including route optimization, collision avoidance, predictive maintenance, and autonomous decision-making. However, its integration also introduces challenges concerning crew roles, legal liability, and regulatory frameworks. Autonomous ships, especially at degrees three and four, lessen the requirement for onboard crew, potentially resulting in job losses or transitions to shore-based positions, igniting ongoing debates and discussions in the maritime industry regarding potential job displacement versus the preservation of traditional seafarer roles (Lee Reference Lee2023).
One of the notable projects involving AI applications in maritime is Yara Birkeland, widely recognized as the world’s first fully autonomous container ship. Developed by a fertilizer company with the support of Rolls-Royce, this container ship utilizes AI for navigation, collision avoidance, and energy management. The ship sails without a crew on board, relying on shore-based control, which is considered the third degree of autonomy (Youd Reference Youd2022). Rolls-Royce’s vision for autonomous vessels demonstrates just how powerful AI can be in achieving fully autonomous operations through ongoing trials (RINA 2018). Forecasts suggest that, by the mid-2030s, up to 3,000 autonomous or semi-autonomous ships will be deployed, necessitating the widespread adoption of AI, according to the HSBA 2018 report. A more recent study conducted by DNV and commissioned by the Maritime Just Transition Task Force predicts that as many as 800,000 seafarers will require additional training and upskilling by the mid-2030s to enable the shipping industry to transition toward alternative fuels and decarbonization (Kaspersen et al. Reference Kaspersen, Karlsen, Helgesen, Giskegjerde, Krugerud and Hoffmann2022). This research highlights the challenges faced by the maritime sector in addressing the increasing uncertainties about the specific skills and profiles required for the future maritime workforce.
3.2.2 Maritime Voyage Optimization
Technologies based on AI systems are revolutionizing maritime voyage planning and route optimization. As one of the most advanced technologies, AI can efficiently process vast amounts of real-time data, such as fluctuations in weather conditions, changing ocean currents, and active ship traffic, to identify the most efficient and secure maritime routes (Riviera News 2022). One example of these technologies is OptiNav AI, a commercialized solution designed as an innovative maritime voyage-planning tool that heavily relies on big data. Powered by cutting-edge technology, this tool strategically utilizes key parameters of maritime operations, including adverse weather conditions, equipment efficiency, and potential security risks.Footnote 3 OptiNav AI, through the use of advanced algorithms, has continually achieved significant fuel savings, demonstrating that such solutions can navigate intelligently – optimizing vessel performance with cost efficiency – while safely navigating the open seas (True North Marine 2025).
AI algorithms also play an important role in conducting accurate energy-based performance analysis of maritime routes. By comparing features such as variable weather conditions, ocean currents, and fuel consumption ratios of ships, the algorithms can optimize energy-efficient routes for vessels (Unoks 2024). With these sophisticated systems, delivery operations can organize power more efficiently, ultimately affecting overall performance and reliability. This technology could simultaneously reduce the need for specific human roles, potentially leading to job displacement. This shift prompts significant debates and discussions about the onboard roles of seafarers, many of whom derive pride and fulfillment from their expertise at sea. However, the integration of AI may lead to the creation of new roles, requiring upskilling and adaptation. As with other emerging technologies, the resilience and adaptability of seafarers will be crucial in aiding navigation through this transformation of the industry.
As this type of technology automates the navigational routines of ships, its impact on seafarers must be carefully managed. The maritime industry should prioritize collaboration between humans and AI over complete automation to maintain human judgment, empathy, teamwork, and creativity.
3.2.3 Predictive Maintenance
AI is also transforming predictive maintenance within the maritime sector, providing substantial improvements in operational efficiency and safety. AI-driven predictive maintenance systems analyze real-time data from sensors installed on ship components, employing machine-learning algorithms to detect early signs of potential failures (Hamidah Reference Hamidah2024). This approach enables ship operators to perform proactive and preventive maintenance, thereby minimizing unanticipated downtime and extending the ship’s lifespan. Through continuous monitoring of equipment health and performance, AI algorithms can identify minor changes in vibration, temperature, or other parameters that may indicate emerging technical issues (Nautilus Shipping 2024). This early detection allows maintenance personnel to address problems before they escalate into major failures, thereby reducing the risks of accidents and costly repairs. A significant advantage of AI-driven predictive maintenance is its ability to integrate and analyze extensive datasets swiftly and accurately, leading to more precise decision-making regarding maintenance schedules (Buzinkay Reference Buzinkay2023).
However, the impact of AI on maritime labor is complex. Although AI-driven predictive maintenance solutions can automate many routine tasks, they are not expected to completely replace crew members. Instead, these machines enhance human capabilities, allowing the crew to focus on more complex and critical responsibilities. By providing real-time data analytics and risk assessments, AI systems empower maritime professionals to make better, informed, and safety-centric decisions (Editorial Team 2024a). Nevertheless, the integration of AI in maritime operations presents challenges. Resistance to change may exist among crew members and other relevant maritime stakeholders, and ensuring data quality and reliability from various sources can be difficult (Durlik et al. Reference Durlik, Miller, Kostecka and Tuński2024).
The deployment of AI systems requires significant investment in technology and training, which may create financial challenges for some maritime operators. Despite these hurdles, the potential benefits of AI in predictive maintenance for maritime work are substantial. By reducing unforeseen failures and improving maintenance schedules, AI helps to create safer conditions for seafarers onboard ships (Nautilus Shipping 2024). Additionally, AI can track the working hours and health indicators of crew members to prevent overexertion and fatigue, common factors in maritime accidents. Moreover, the growing integration of AI technologies in the maritime industry for predictive maintenance is expected to change the roles of the workforce onboard ships. While some tasks may become automated, new roles will arise, concentrated on supervising and evaluating AI-generated insights. This transformation will demand ongoing training and upskilling to ensure effective collaboration with AI systems (Editorial Team 2024a).
3.2.4 Maritime Security and Risk Management
AI applications in maritime security and risk management cover several areas, including predictive maintenance, route optimization, cybersecurity, and port operations. AI-driven sensors and machine-learning algorithms facilitate operations in turbulent seas.
The legal framework governing AI in maritime operations continues to evolve in order to address the unique challenges posed by this technology. The IMO has recognized the importance of cybersecurity in maritime operations and has released guidelines for protecting digital systems on ships. However, the implementation of these cybersecurity standards for MASS presents new challenges, as their reliance on AI and external communication systems increases their susceptibility to cyber threats.
A key legal concern is liability in mishaps utilizing AI-driven systems. The conventional notion of flag State responsibility, as outlined in the United Nations Convention on the Law of the Sea,Footnote 4 may need reassessment in relation to AI-operated vessels. For instance, if a system failure leads to a multi-vessel collision in international waters, establishing liability becomes complicated when several ships rely on the same machine-learning algorithms (Adnan Reference Adnan2023). The EU’s Artificial Intelligence Act (AI Act),Footnote 5 currently in development, is anticipated to significantly impact the maritime sector. This legislation aims to regulate AI systems based on their risk levels, affecting all areas of the shipping industry. The AI Act underscores safety, data protection, transparency, and accountability tied to the use of AI systems, which will certainly shape the design and operation of autonomous vessels.
Incorporating AI into maritime operations offers advantages and challenges for seafarers. While AI can improve safety by minimizing human errors and offering real-time decision support, it raises concerns about job displacement. Automating maritime operations may result in considerable job displacement among seafarers, requiring a transformation in skills and roles within the industry.
The legal ramifications of AI in maritime security pertain to data protection and privacy. MASS produces and conveys substantial volumes of data, encompassing sensitive details regarding vessel operations and positioning. Securing this data is essential to avert unauthorized access and privacy violations. Therefore, a comprehensive legislative framework is essential to govern the collection, storage, and transfer of data from ships, emphasizing protection against cyberattacks.
3.2.5 Use of AI in Employment Management
While automation provides numerous advantages for managing seafarers’ employment, it also presents notable challenges. The incorporation of process automation and AI into the maritime industry can greatly diminish labor demand, particularly for low-skilled positions. However, highly skilled roles, such as captains and engineers, are likely to be less impacted (WMU 2023).
To mitigate the impact of automation on employment, it is crucial to invest in training, upskilling, and reskilling seafarers while also ensuring their retention in the sector. Improving seafarer qualifications and training seafarers in emerging technologies are essential to help the workforce adapt to the sector’s demands. At the European level, for example, Recommendation (EU) 2024/236 of the European Commission emphasizes the importance of skill-based training.Footnote 6 This training should focus on knowledge, methodology, participation, and adapting workers’ skills to new technologies and processes. Government training initiatives and the modernization of regulated vocational training are key tools in preparing maritime professionals for the automation era. This effort can transform challenges associated with automation into a productive advantage rather than a threat. AI-driven tools can further support the analysis of future skill needs and gaps at the national and industry levels and help individuals identify potential career paths and learning opportunities. These tools, which are already utilized by some public employment services (OECD 2022), can also offer opportunities for additional systemic solutions in various transportation sectors, not just maritime.
3.3 Implications of AI on the Development of Maritime Labor Laws
Multiple scientific papers have reported that the integration of AI in maritime shipping, particularly in the development of autonomous ships, is expected to reduce crew sizes on board, which may create fewer job opportunities. However, it could also generate new positions, such as remote operators that require specialized skills, advanced training, and technical competencies. These developments may necessitate revisions to labor laws to ensure clear and fair working conditions for the maritime workforce, including seafarers, while also addressing legal liability in accidents involving AI systems. The current legal framework, specifically the MLC, protects formal seafarers’ rights on ships but does not clearly address how those rights apply to unmanned vessels, potentially necessitating legislative amendments to the Convention. Additionally, other key instruments, such as the STCW, also require revision and adaptation to the new paradigm of technology use in the maritime sector. As AI and automation technologies continue to reshape the maritime industry, legal and institutional structures must evolve accordingly to safeguard seafarers’ rights, redefine new roles, and, importantly, establish clear accountability in AI-driven maritime operations.
3.3.1 Ongoing Discussions and Future Outlook
International organizations, including the IMO, are endeavoring to regulate autonomous vessels, taking into account labor law ramifications, including the definition of “crew” in the context of remote operations. The International Labor Organization (ILO), although not explicitly addressing autonomous ships, is examining the larger labor implications of AI, which may guide the maritime sector adjustments. Upcoming modifications may entail revised training mandates and alterations to social security, mirroring the dynamic characteristics of maritime employment that are likely to influence different maritime legal instruments governing seafarers’ rights.
3.3.1.1 Maritime Labor Instruments
The MLC, adopted by the ILO, is a comprehensive convention that sets minimum standards for the working and living conditions of seafarers. It entered into force on August 20, 2013, and includes aspects such as minimum age, medical certification, employment agreements, wages, hours of work and rest, leave, repatriation, social security, health and safety, accident prevention, and on-board medical care, among other fundamental labor rights for seafarers. Its primary goal is to ensure the fair employment of seafarers and equitable competition among shipowners, applying to ships entering the ports of ratifying States and those flying their flag. The MLC is not the only regulatory framework governing maritime labor. Other significant instruments adopted by the IMO, such as the STCW, address training and certification relevant to the emerging skill requirements in an AI-driven environment. The STCW was initially developed in 1978 in response to the Torrey Canyon disaster in 1967. The Convention has been continuously revised, with major amendments in 1995 and 2010 following diplomatic conferences. To date, 195 modifications have been introduced to the STCW Code. The Code outlines the basic competency levels for several maritime professions within a series of tables in its mandatory Part A; its nonbinding Part B offers recommendations for the effective implementation of the Convention.
Despite their significance in maritime labor law, these instruments do not yet address the complexity and challenges of the human elements associated with autonomous vessels. Challenges for maritime law regarding the integration of AI (primarily autonomous ships) include, but are not limited to, crew size reduction (and potential job loss), new roles and proficiencies, working conditions for remote operators, legal responsibility and liability, and social security and benefits.
Potential Reduction in Crew Size and Job Implications.
Autonomous ships could significantly reduce the need for onboard crew, leading to job displacement or transformation. A study conducted by Kretschmann et al. highlights the cost advantages of autonomous ships through decreased crew expenses, but also points out potential increases in port expenses and monitoring costs (Nguyen et al. Reference Nguyen, Ruzaeva, Góez and Guajardo2022). Discussions in both industry and academia reflect this dual perspective. For instance, the World Maritime University report on the future of work indicates that whether autonomous ships will signal the end for seafarers’ jobs or not, there will still be a need for seafarers to oversee and maintain these ships when necessary, with appropriate training and skills (WMU 2023). In contrast, as mentioned earlier, another study asserts that seafarers’ jobs will continue to exist even with the increased development of autonomous or semi-autonomous ships, albeit in evolving and more complex capacities and roles (Nautilus International 2018). This controversy underscores the need for international and national labor regulations to address seafarers’ employment security and retraining. In this context, the provisions regarding wages and employment agreements in the MLC may become less relevant for crewless ships, necessitating legislative amendments to support shore-based roles or alternative employment for maritime professionals adapting to the new technological landscape shaped by AI and automation.
New Roles and Skills.
The impact of AI on maritime operations creates employment opportunities for remote operators who oversee autonomous ships from shore. The interim requirements for MASS trials established by the IMO and confirmed in 2019 underscore the essential need for a maritime workforce – which includes all maritime professionals operating on board, on shore, or remotely – to possess adequate qualifications and experience (IMO 2019). Such a shift necessitates revisions to the STCW, ensuring that training programs incorporate skills in AI system administration, as well as remote operations such as cybersecurity. Equipping seafarers with these skills will enable them to safely transition into these new roles. This transition is crucial not only to secure seafarers’ employment but also to ensure that labor regulations serve a protective role in substance. In 2017, the IMO initiated a regulatory scoping exercise (RSE) to evaluate how the organization’s legal instruments could support MASS operations. The RSE identifies the STCW and its Code as critical focus areas for the IMO, indicating that substantial efforts must be made to provide necessary clarifications and guidance before implementing MASS operations with advanced degrees of autonomy on a broader scale. The results of the RSE are documented in MSC.1/Circ.1638 (IMO 2021). The STCW outlines three primary components: (1) definitions and clarifications regarding the roles of the master, crew, and responsible individuals; (2) definitions and clarifications relating to remote control centers; and (3) definitions and clarifications concerning remote operators classified as seafarers. The RSE primarily stressed the importance of defining the new responsibilities of the crew onboard ships and remote operators. It emphasized the need to clarify the connection between remote operators and onboard personnel. However, the RSE did not specify the necessary capabilities for crew members or remote operators involved in MASS operations. Thus, modifications to the STCW and its Code should be evaluated with regard to new technologies or automated procedures. The organization’s structure and the roles of different parties in implementing MASS, along with the required competencies to undertake these additional tasks, remain unclear. Consequently, the maritime community currently faces challenges in determining new training requirements due to the industry’s limited expertise with MASS.
It is important to note that developing new educational and training programs requires significant time and resources. Training and education for seafarers are unique because they include a professional qualification, as outlined by the STCW Code, while also integrating an academic component, which in many countries is incorporated into the higher education system. This duality results in the attainment of a professional certification, specifically a Certificate of Competency, along with an academic degree upon successfully completing the relevant academic programs. Higher education institutions must justify any changes to their academic programs to provide relevant education. Given the considerable time gap between developing an academic program and the subsequent entry of graduates into the labor market, further discussions are essential regarding the curriculum of relevant academic programs and the methods of training and education. The STCW, in its current form, lacks the necessary training materials and organization required for the operation of MASS. In this regard, only a limited number of research projects have begun efforts to investigate the proper training requirements for maritime operations involving MASS ships and other emerging technologies.Footnote 7
Beyond Training: Continuous Professional Development for Seafarers.
As ships evolve from manned to remotely operated or fully autonomous, the responsibilities of ship operators will shift from direct navigation to managing supervisory autonomy. The required expertise includes algorithmic interpretability, human–autonomy collaboration, and crisis response, which may become the primary training objectives. In the future, remotely operated and autonomous ships will need to coexist with conventional vessels, making the roles of masters and engineers crucial for operational continuity, safety, and regulatory compliance (Kim and Schröder-Hinrichs Reference Kim, Schröder-Hinrichs, Ko and Song2021). The diverse landscape of global shipping, characterized by the coexistence of autonomous, remotely operated, and crewed vessels in shared navigational areas, demands that maritime professionals possess both traditional seafaring skills and advanced technical competencies to manage effectively the interactions among these various types of vessels.
New roles are expected to emerge, including remote operation operators, AI ship engineers, and maritime cybersecurity managers, to align with future demands. The integration of AI, automation, and remote operations in shipping is expected to transform numerous professions in the maritime industry, transitioning career paths from conventional hazardous and labor-intensive positions to highly specialized and technology-driven roles. This will augment the intellectual and strategic sophistication of maritime vocations, boost welfare and working conditions, and elevate their status to a level commensurate with specialists in aerospace, robotics, and high-tech industries. In this context, conventional seafarer career trajectories, defined by a hierarchical structure (that is, rank-based) and flexibility (that is, employment per voyage contract), are poised for transformation.
Health, Safety, and Social Security Protection.
The transformation of crew structure due to AI-driven automation will impact social security and benefits. Seafarers are entitled to benefits if their employment terms are impacted by technological development or work performed onshore. However, according to the provisions of the MLC, their benefits, such as medical care and repatriation, may need to be reevaluated. Therefore, inclusive policies are necessary to ensure the welfare of seafarers in an AI-driven future (Kim and Schröder-Hinrichs Reference Kim, Schröder-Hinrichs, Ko and Song2021). The influence of AI on health and safety in maritime operations is substantial, particularly as autonomous ships reduce the need for onboard seafarers. Remote operators may experience isolation or stress due to constant surveillance, and onboard crews of semi-autonomous ships must contend with AI malfunctions, which increase the likelihood of accidents. The MLC’s health and safety provisions, specifically accident prevention under Title 4, require modifications to address these issues, including extending coverage to cover shore-based roles. Social protection, including medical care and repatriation under the MLC, may need reassessment for reduced crew size or shore-based personnel. For its part, the STCW must ensure that seafarers are adequately trained to manage these new situations. Cybersecurity training, for instance, becomes crucial due to the digital nature of AI and other emerging technologies (IMO 2025b). Social protection may guide the STCW’s long-term welfare objectives, ensuring that training includes emergency response for AI-related incidents and cyberattacks.
To address these challenges, tripartite collaboration among governments, shipowners, and representatives of seafarers is essential. In the maritime sector, this approach could influence not only the MLC, which is part of its functionality, but also revisions to the STCW. The IMO’s 2021 regulatory scoping exercise for MASS identified gaps in labor law, including issues related to training and certification, suggesting that tripartite dialogue is necessary to define “crew” and “responsible person” for remote operations (IMO 2025a). For the STCW, tripartite engagement could establish new competency standards for AI interaction, data analysis, and cybersecurity. Social partners, such as the International Transport Workers Federation and Nautilus International, play a critical role in shaping training requirements and aligning them with industry needs and technological advancements.
Legal Responsibility and Liability.
Determining liability in accidents involving autonomous ships is a complex issue. Traditional labor regulations assign responsibility and accountability to the ship’s master and crew, but with AI, these could shift to shipowners, AI technology developers, software developers, or remote operators. This uncertainty has implications for insurance providers, which must develop new models to cover the unique risks associated with AI. The IMO’s legal committee scoping exercise, as part of the 2021 regulatory review, identifies gaps in liability and compensation treaties, suggesting a need for new frameworks (IMO 2025a). This affects labor law, particularly concerning compensation for injuries or incidents, potentially requiring amendments to the MLC’s provisions on accident prevention and shipowner liability.
3.3.1.2 Working Conditions and Potential Control of Workers at Sea
Numerous studies have linked electronic or digital monitoring systems with various aspects of working conditions, considering factors such as employee performance, job duration, work rhythms, collaboration between employees and employers, skill development, and employee well-being. Human capital management and performance management are two primary human resources functions enhanced by AI, though they come with implications for occupational safety and health (OSH). “People analytics” solutions are recognized for improving the recruitment process by enabling informed decisions regarding potential employees, based on their access to relevant data. Certain types of systems allow employers to conduct video interviews to reduce specific biases during the interview process. They can also assess employee performance through performance-based compensation to aid in formulating corporate strategies for individual workers. Without human oversight and ethical considerations, this algorithmic decision-making tool has shown significant potential to create OSH issues, including increased structural, physical, and psychosocial risks such as stress and anxiety (Moore Reference Moore2019, 93). Employees may be led to question the fairness, integrity, and accuracy of decisions made by their employers – such as job displacement and workplace restructuring – due to their lack of access to data derived from people analytics systems.
Workers may feel that they are being monitored if they think that people analytics data is used for performance management without appropriate oversight. This perception, along with the fear of layoffs resulting from performance evaluations, may drive them to increase their work effort excessively due to job loss anxiety, potentially leading to OSH concerns. The most urgent issues related to AI in workplaces stem from using people analytics to gather employee data for decision-making. Other AI systems in the workplace that have contributed to OSH concerns include cobots and chatbots. These AI-enhanced tools have been integrated into various industrial processes, causing psychosocial issues linked to the fear of job displacement due to automation. Different sectors, including automotive manufacturing, utilize cobots to perform tasks that would typically require significantly more human time. Robotic arms have replaced human hands, and AI has enhanced robots with cognitive abilities, allowing them to think like humans and effectively take over human intellect (Moore Reference Moore and De Stefano2020).
A report indicates that using cobots has reduced OSH risks by lowering workers’ exposure to hazardous ergonomic, physical, and chemical conditions (Moore Reference Moore2019, 93). Conversely, another study has identified three primary categories of OSH issues arising from interactions among humans, cobots, and the environment: (1) collision hazards between robots and humans arising from unexpected robot behavior due to machine learning; (2) security vulnerabilities stemming from the robot’s internet connectivity, which can compromise software integrity and expose the system to threats; and (3) environmental risks, where unpredictable human actions and sensor degradation in unstructured environments can lead to hazards (Moore Reference Moore2019, 93). Moreover, AI has facilitated the integration of voice recognition and machine vision into chatbots, which is recognized to threaten both unskilled and skilled jobs due to the increasing automation of tasks traditionally performed by humans. For example, in a chemical firm that manufactures optical peripherals for machinery, an individual would spend numerous hours examining repetitive images of tiny machine chips to detect errors. AI has replaced human labor in this task, eliminating occupational health issues such as musculoskeletal disorders and visual strain. However, improperly utilized AI-augmented robots in factories may induce stress among workers. It is noted that integrating automation, algorithmic management, and digitalization can create a harmful system that may result in psychosocial problems, particularly when employees are expected to perform at a robotic pace instead of allowing robots to operate at a human pace. Occasionally, an employee monitors a single machine that sends notifications to their electronic devices, such as smartphones or computers. This can lead to other psychosocial problems, such as job overload, where individuals continue to work beyond their designated hours.
At sea, the ship is limited in space and vulnerable to navigational hazards. Employment onboard ships is influenced not only by potential accident risks but also by factors of movement and isolation. Life on board requires the integration of work and personal life in a single environment, highlighting the importance of work management along with the composition and size of the crew. Improving onboard working conditions relates to the quality of the ship’s habitability and the safety of life at sea. Maritime surveillance may involve both security and control, as seafarers are required to follow the directives of the shipowner through the ship’s master. As a result, surveillance techniques serve various purposes, including safety, security, regulatory compliance monitoring, and the enhancement of commercial operations. The tripartite objective of overseeing the vessel and its crew is a responsibility of shipowners established by international conventions set by the IMO and the ILO.
Emerging technologies such as AI onboard ships can also increase labor intensity through continuous surveillance and inconsistent scheduling (De Stefano Reference De Stefano2018). In maritime operations, remote operators supervising autonomous vessels may face similar pressures as AI systems control duties and working hours. The MLC’s provisions on work hours and rest for onboard seafarers may not directly apply to shore-based positions, necessitating new regulations to ensure fair treatment, including rights to disconnect and protection from excessive monitoring, in line with general labor regulation trends. Therefore, seafarers have the right to negotiate the use of algorithms and to oversee big data onboard ships. Seafarers’ unions should advocate for safeguards against excessive surveillance of seafarers through various technologies and ensure equitable working conditions (Lagdami Reference Lagdami2023).
Therefore, to ensure the responsible and ethical use of AI aboard ships, it is crucial to focus on four key areas. First, seafarers should have ownership and control over data produced during their work on ships. This principle recognizes the importance of individual privacy and data rights in an increasingly digital maritime environment. Allowing seafarers to control their data fosters trust and transparency in the application of emerging technologies at sea. Second, data generated from seafarers should only be used for purposes related to maritime safety and security. This information should be limited to protect the privacy and interests of maritime workers and to focus on enhancing safety and security in maritime activities. Clearly defining how the data will be used will help prevent its misappropriation or abuse. Third, international regulations should govern the use of AI on board, enshrining the principle of ultimate human responsibility for its impacts. Fourth, seafarers should be trained to address new challenges related to emerging technologies. Seafarers should be aware of the technical, legal, economic, and ethical issues associated with the use of AI-based tools (Lagdami Reference Lagdami2023).
3.3.1.3 Cybersecurity and the Leakage of Seafarers’ Personal Information
The integration of AI and big data in the maritime industry also increases vulnerability to cybersecurity breaches. As we transition more seriously into the digital era, cybersecurity breaches have surged, with operational technology attacks soaring by as much as 900 percent in the past three years (Akpan et al. Reference Akpan, Bendiab, Shiaeles, Karamperidis and Michaloliakos2022). Systems that rely on interconnected networks, cloud platforms, and real-time data exchange become prime targets for cyberattacks. Various threat actors could exploit these vulnerabilities to access information and manipulate or disrupt critical operations at sea. For instance, a cyberattack on an AI-powered navigation system could result in an incorrect route for vessels, thereby putting crew and cargo in dangerous situations. Likewise, a breach in big data systems can adversely affect operations at ports or disrupt global supply chains (Kanellopoulos Reference Kanellopoulos2024).
Notable incidents have been demonstrated in the outcomes of such breaches – for example, the ransomware attack on Maersk, the largest shipping company in the world, in 2017. This incident was essentially operational, except for the broader implications on data security; sensitive information and operational data were at risk of corruption during the rapid recovery process (Clavijo Mesa et al. Reference Clavijo Mesa, Patino-Rodriguez and Guevara Carazas2024). Similarly, in 2020, the Mediterranean Shipping Company encountered a cyber-related incident involving unauthorized access to its customer database (Ben Farah et al. Reference Farah, Amine, Ukwandu, Hindy, Brosset, Bures, Andonovic and Bellekens2022). This breach disrupted container bookings and exposed sensitive data, undermining customer trust and necessitating expensive mitigation efforts. These incidents have underscored the growing need for robust cybersecurity measures to safeguard sensitive information – including seafarers’ personal details – from theft and misuse.
By far the most concerning impact of such incidents is the leakage of seafarers’ personal information in the era of digitalization onboard ships. In fact, with the increasing use of networked digital tools in everyday operations, seafarers are particularly at risk from hacking incidents (Finn Reference Finn2020). In addition, current maritime education and training programs lack a proper element of cybersecurity, rendering seafarers poorly prepared while on the job to identify and neutralize any cyber threats (Heering Reference Heering2020). Such cyber incidents put at risk sensitive information such as medical records, identity details, and financial data. In addition, such information may lead to identity theft, fraud involving financial transactions, and psychological trauma, among other impacts, which can further increase the issue of the criminalization of seafarers. Trust between seafarers and their employers is also undermined whenever inadequate steps have been taken to secure information. These incidents could also cause profound reputational harm to shipping companies, leading to possible legal liabilities, regulatory penalties, and the erosion of stakeholder confidence and trust. Therefore, maritime cybersecurity risks must be dealt with through robust regulatory frameworks and international cooperation to keep sensitive data secure and ensure the resilience of maritime operations.
3.4 The AI Act from a Maritime Labor Perspective
The AI Act – which was adopted in June 2024, entered into force on August 1 of that year, and became fully applicable by August 2026 – is considered the first ever regulatory legal framework governing AI and its challenges and risks. Thus, it puts Europe at the forefront of international AI regulation.
The AI Act was developed based on a risk model; it in fact clusters AI systems according to the risks posed to humans and society (Kosinski and Scapicchio Reference Kosinski and Scapicchio2024). The Act categorizes AI systems into four risk levels: unacceptable risk (prohibited), high risk (strict requirements), limited risk (transparency obligations), and minimal or no risk (no specific obligations; European Parliament 2025). High-risk systems, such as those used in critical applications like medical devices or transportation, must comply with requirements including risk management, data governance, transparency, and human oversight (Meier and Spichiger Reference Meier and Spichiger2024). The Act prohibits certain AI practices, imposes strict regulations on high-risk AI systems, and sets transparency requirements for limited-risk AI applications. This approach permits a nuanced application of rules, imposing stricter regulations that should be applied to high-risk AI systems and less stringent rules for those deemed lower risks. The Act explicitly bans numerous AI methods, especially those that lever human behavior or exploit specific vulnerable groups. Thus, the AI Act complements the General Data Protection Regulation (GDPR)Footnote 8 with its governance framework. It strengthens some of the principles of the GDPR, such as the lawfulness of processing personal data, purpose limitation, and transparency. The Act affirms that AI systems must comply with the same legal basis for processing information under the GDPR; thus, no data can be collected for wrongful purposes (Meier and Spichiger Reference Meier and Spichiger2024). It also sets out specific rules for protecting personal data in AI systems. The Act permits the processing of sensitive personal data for bias detection and mitigation in high-risk AI systems, but only under stringent conditions (Hullen Reference Hullen2024). This is one of the conditions necessary to ensure nondiscrimination and fairness in AI applications. In addition, the Act emphasizes data minimization – requiring AI systems to use only the data necessary for the given purpose, which is aligned with the GDPR and reduces the risk of data breaches or misuse. In a move toward such a new legal environment, the European Commission launched the AI Pact, a voluntary scheme to encourage AI developers in and out of European geographies to implement key provisions of the Act in preparation for the legal timeline. In its first expression of interest in the AI Pact, in November 2023, over 550 organizations from various geographies, sectors, and sizes demonstrated a strong willingness to comply proactively.
3.4.1 Intersection of the AI Act with the MASS Code
Currently, the IMO is advancing the development of the MASS Code, which serves as a regulatory instrument for MASS capable of operating at various levels of autonomy, as discussed earlier. The nonmandatory MASS Code aims to ensure safety, security, and environmental protection, and it was expected to be adopted by May 2025. In contrast, a mandatory code could come into effect as early as July 2030, with enforcement potentially beginning in January 2032 (Editorial Team 2024c). This framework is essential as a wide range of sectors within the marine industry start to implement AI and automation, which could transform traditional roles and responsibilities.
The AI Act significantly affects the maritime industry, particularly in relation to MASS. The Act establishes a risk-based framework for AI regulation, which has real implications for autonomous ships where AI systems are responsible for navigation, collision avoidance, and other essential functions. This classification necessitates rigorous testing for compliance before these systems can be commercialized and for subsequent adherence to operational and safety standards for autonomous vessels. On a broader scale, the Act requires human oversight for high-risk AI systems, which could entail ensuring remote monitoring for fully autonomous MASS (degree four) when no crew members are present. The MASS Code, centered on safety, likely aligns by mandating human intervention capabilities, especially for higher levels of autonomy, to address risks such as cyberattacks or system failures (Kepesedi Reference Kepesedi2022).
The AI Act’s requirements for high-risk AI systems will likely influence the MASS Code, particularly for MASS operating in or related to the EU. For degrees one and two, human oversight is clear, with seafarers on board providing direct control. However, for degree three, remote operators must ensure effective monitoring, aligning with the Act’s transparency and oversight requirements. Degree four, which involves fully autonomous ships, presents a significant challenge, as the Act mandates human involvement, potentially requiring remote monitoring systems or contingency plans for human intervention (Lölfing Reference Lölfing2023). As the MASS Code is still under development, it must incorporate these requirements to ensure compliance, potentially leading to provisions for remote operation centers and clear definitions of remote operator responsibilities (Editorial Team 2024b).
So far, there are no examples directly linked to the AI Act and MASS; however, parallels can be drawn from regulations on autonomous vehicles. For instance, the Act’s influence on autonomous vehicles indicates that MASS providers might benefit from AI regulatory sandboxes, which would allow testing in controlled environments (Güçlütürk and Vural Reference Güçlütürk and Vural2024). Trials of MASS conducted by companies in Korea, Japan, and Norway highlight the necessity for robust AI systems with human oversight, aligning with both frameworks.
3.4.2 The AI Act’s Potential Impact on Seafarers’ Roles and Job Prospects
As previously discussed, the AI Act requires human involvement for high-risk AI systems, including MASS, for navigation and decision-making. Even fully autonomous ships (degree four, with no crew on board) may need remote monitoring or intervention, shifting traditional seafaring roles to shore-based positions such as operators based in remote centers. For ships with lower autonomy (degrees one and two, with crew on board), seafarers will continue to oversee AI systems. Still, their roles may evolve to include more monitoring and less hands-on control. This change could reduce demand for on-board seafarers, potentially leading to job displacement, but it also creates new opportunities in remote operation centers. As a general fact, seafarers may need retraining and upskilling to adapt to these new roles, focusing on competencies such as remote monitoring, AI system management, and cybersecurity. This aligns with the AI Act’s focus on a human-centric AI approach and may involve collaborations with training institutions both inside and outside Europe (Meier and Spichiger Reference Meier and Spichiger2024).
3.4.3 Potential Opportunities and Challenges
The alignment of the MASS Code with the AI Act could enhance safety. However, it faces challenges – such as defining human oversight for fully autonomous ships, harmonizing regulatory frameworks, and addressing potential cyber threats – as highlighted in a UN Trade and Development report (Kepesedi Reference Kepesedi2022). Unlike conventional ships, where crew members can respond immediately to emergencies, remote operators must rely on robust communication systems and real-time data to make critical decisions. This dependence on technology introduces new vulnerabilities, such as cybersecurity risks and threats, along with the possibility of communication breakdowns, which could compromise the safety of the vessel, its cargo, and the marine environment. Solutions may include remote monitoring systems or innovative contingency plans. This intersection emphasizes the need for international cooperation, such as between IMO and EU bodies, and the development of codes of best practice for AI in maritime applications (Caroli Reference Caroli2025).
The future outlook suggests a balanced approach that ensures innovation while prioritizing safety and ethical considerations, with ongoing updates to the AI Act and MASS Code as technology evolves. This intersection also underscores the need for collaborative regulatory efforts to navigate the complexities of autonomous shipping and the use of emerging technologies, such as AI, onboard ships. The development of best practice codes for AI in maritime applications is another area where international cooperation could yield significant benefits. Such practice codes could standardize training requirements and define roles for seafarers involved in remote operations, ensuring a consistent approach across different jurisdictions (Wylie Reference Wylie2024). This standardization is crucial for building trust in the use of autonomous shipping technologies and facilitating their global adoption.
Last but not least, as the maritime industry navigates these uncharted waters, it is clear that successfully integrating emerging technologies such as AI into global shipping operations will require a delicate balance between technological innovation and regulatory oversight. The ongoing dialogue between regulatory bodies, industry stakeholders, and technology providers will be crucial in shaping a future where autonomous ships can operate safely and efficiently alongside conventional vessels.
3.5 Conclusion
AI is increasingly being integrated into the maritime domain, driving a profound shift across the operational efficiencies, technology, human element, and regulatory frameworks. Through maritime AI application solutions, including voyage optimization, significant improvements in safety, economic effectiveness, and environmental sustainability can be achieved. Such shifts, however, must be approached with caution, particularly their impact on seafarers and the legal frameworks governing their employment and responsibilities.
At the core of this transformation is the evolving nature of seafarers’ roles and the competencies they must acquire and master. While concerns over job loss persist, research indicates that AI is more likely to transform seafarers’ roles rather than eliminate them, which will increase the demand for new competencies such as data analysis, monitoring AI systems, and operating more complex systems. This shift necessitates a proactive approach to maritime education and training, ensuring that the maritime workforce is able to collaborate with AI systems in tasks and adapt to the changing demands of the industry.
The growing use of AI in maritime operations raises significant legal and ethical concerns. Solutions are needed to address concerns related to liability and accountability that arise from AI system accidents, along with continuing concerns over data privacy and cybersecurity and the need for global coherence among technical standards. Current legal frameworks, largely designed for traditional maritime operations, may not adequately address the unique challenges posed by AI, thus requiring careful scrutiny and realignment to ensure clarity and legal certainty.
The IMO has initiated key efforts to address these challenges, such as developing a strategy on e-navigation and working groups on MASS. These initiatives are essential to establishing a consistent regulatory framework. The initiatives call for more cooperation between different industry actors to develop regulations that balance the need to accelerate AI adoption in the maritime industry with the need for caution and prudence regarding how this technology may affect the global maritime workforce.
While AI-driven transformation represents unprecedented opportunities in the maritime industry, its ultimate success can be achieved only through a holistic solution that addresses human factor concerns and legal uncertainties and promotes multinational cooperation. Innovation in AI is only valuable in the maritime industry when it upholds fundamental principles of safety, security, and occupational health and well-being. The maritime industry must ensure that new technologies complement the work of maritime professionals rather than eliminate it. The responsible proliferation of AI in the maritime sector necessitates a continuous learning mindset, fostering an adaptability and knowledge-sharing culture. Furthermore, prioritizing ethical AI deployment is of prime importance. While ensuring regulatory clarity and workforce development, the maritime industry can efficiently leverage automation and AI’s potential benefits to ensure sustainable and equitable technological advancement.
Open access