Skip to main content
  1. AI Standard of Care by Industry/

Maritime & Shipping AI Standard of Care

Table of Contents

The maritime industry stands at the threshold of its greatest transformation since containerization. Autonomous vessels are transitioning from concept to commercial reality, with fully unmanned cargo ships now operating in controlled waterways and semi-autonomous systems augmenting bridge crews worldwide. Port operations increasingly rely on AI for everything from crane control to vessel scheduling to security monitoring.

Yet maritime AI presents unique liability challenges unlike any other transportation sector. Ships operate across international boundaries, subject to overlapping national and international regulatory frameworks. Ancient maritime law principles, general average, limitation of liability, salvage rights, must now accommodate algorithmic decision-making. When an AI collision avoidance system makes a fatal navigation error, which nation’s courts have jurisdiction, what law applies, and who bears responsibility?

$150B+
Maritime AI Market
Projected by 2030
70%
Ship Incidents
Attributed to human error
2028
Target Date
IMO MASS Code implementation
400+
Autonomous Vessels
In operation or development

The Emergence of Maritime Autonomous Surface Ships (MASS)
#

Levels of Autonomy
#

The International Maritime Organization (IMO) has defined degrees of autonomy for Maritime Autonomous Surface Ships (MASS):

DegreeDescriptionHuman Role
Degree 1Ship with automated processes and decision supportSeafarers on board control systems
Degree 2Remotely controlled ship with seafarers on boardShore-based operators can control
Degree 3Remotely controlled ship without seafarersShore-based operators control fully
Degree 4Fully autonomous shipShip makes decisions independently

Current Autonomous Vessel Deployments
#

Autonomous maritime technology is advancing rapidly:

Operational Autonomous/Remote Vessels:

  • Yara Birkeland (Norway): First fully electric autonomous container ship
  • Suzaku (Japan): Autonomous coastal freighter
  • Prism Courage (Korea): First autonomous ocean crossing (2022)
  • Mayflower Autonomous Ship: Transatlantic research vessel

Semi-Autonomous Systems in Wide Use:

  • Dynamic positioning systems
  • Autonomous collision avoidance aids
  • Automated mooring systems
  • AI-assisted navigation plotting
The Remote Operations Model
Rather than fully autonomous operation, many shipping companies are pursuing the remote operations model: ships with minimal or no crew controlled from shore-based operations centers. This approach maintains human decision-making while dramatically reducing crewing costs. Norway’s Massterly operates the world’s first autonomous ship operations center, demonstrating the viability of shore-based vessel control. However, this model raises critical questions about situational awareness, response time, and the standard of care for remote operators.

IMO Regulatory Framework for MASS
#

MASS Code Development
#

The International Maritime Organization has been developing a comprehensive framework for autonomous vessels:

Timeline:

  • 2017: IMO begins regulatory scoping exercise for MASS
  • 2019: Interim guidelines for MASS trials issued
  • 2021: Maritime Safety Committee (MSC) continues instrument review
  • 2024: Draft MASS Code developed
  • 2025: MASS Code adoption expected
  • 2028: Target implementation date

Key IMO Principles for MASS
#

The emerging IMO framework establishes:

Equivalence Principle:

  • MASS must achieve safety level at least equivalent to conventional ships
  • Autonomous systems must not introduce new unacceptable risks
  • Technology must not diminish environmental protection

Human Element:

  • Human oversight must be maintained at appropriate level
  • Remote operators require equivalent competency to shipboard officers
  • Clear allocation of responsibility between human and machine

Flag State Responsibility:

  • Flag states remain responsible for MASS flying their flag
  • Certification and inspection requirements apply to autonomous features
  • Flag states must ensure adequate oversight capability
Flags of Convenience and MASS
The MASS Code will only be as effective as its enforcement. Maritime’s long history of flag-of-convenience arrangements, where ships register in nations with minimal oversight, raises concerns about autonomous vessel safety. Will Panama, Liberia, and Marshall Islands (which together flag over 40% of world tonnage) effectively oversee autonomous fleets? The liability implications for underregulated autonomous ships are significant.

COLREGS and Autonomous Navigation
#

The International Regulations for Preventing Collisions at Sea (COLREGS) present fundamental challenges for autonomous vessels:

Key COLREGS Requirements:

  • Rule 2: “Ordinary practice of seamen” judgment calls
  • Rule 5: Proper lookout “by all available means”
  • Rule 7: Use of “all available means appropriate to the prevailing circumstances”
  • Rule 8: Action to avoid collision “in ample time… with due regard to good seamanship”

The Problem: COLREGS assume human judgment, seamanship, and the ability to assess “ordinary practice.” Can an AI system exercise “good seamanship”? When COLREGS require action that would be “apparent to the ordinary practice of seamen,” what standard applies to algorithms?

The IMO is developing interpretive guidance, but fundamental questions remain about how AI can comply with rules designed for human mariners.


Port Automation and AI
#

Automated Terminal Operations
#

Major ports worldwide have implemented extensive automation:

Automated Container Terminals:

  • Rotterdam Maasvlakte II: Fully automated stacking and horizontal transport
  • Qingdao Qianwan: Largest automated terminal in Asia
  • Long Beach Middle Harbor: Most automated US terminal
  • Singapore Tuas: Next-generation mega-automated port

AI Applications in Ports:

  • Automated ship-to-shore cranes
  • Autonomous guided vehicles (AGVs) and straddle carriers
  • AI-optimized vessel scheduling and berth allocation
  • Predictive maintenance for port equipment
  • Automated gate systems and security screening

Port Automation Incidents
#

Automated ports have experienced significant incidents:

YearPortIncidentOutcome
2016Los AngelesAutomated crane dropped containerWorker fatality
2018RotterdamAGV collision during testingEquipment damage
2020SingaporeAutomated rail system malfunctionOperations shutdown
2022HamburgAI scheduling system failureVessel delays
2023AsiaMultiple AGV near-missesSafety review
Labor and Safety Tensions
Port automation has generated significant labor disputes, with unions arguing that reducing human workers increases safety risks. The International Longshore and Warehouse Union (ILWU) and International Dockworkers Council have both raised concerns about automated terminal safety. While automation proponents cite reduced human exposure to hazards, critics argue that remaining workers face increased risks from human-machine interaction failures.

Collision Avoidance and Navigation AI
#

AI-Enhanced Bridge Systems
#

Modern vessels increasingly rely on AI for navigation safety:

Systems in Use:

  • Electronic Chart Display and Information Systems (ECDIS) with AI enhancement
  • Automatic Identification System (AIS) analysis and prediction
  • AI-powered collision avoidance recommendations
  • Weather routing optimization
  • Dynamic under-keel clearance systems

Collision Avoidance System Failures
#

AI navigation aids have been implicated in maritime incidents:

Documented Issues:

  • Overreliance on ECDIS contributing to groundings
  • AIS spoofing and data integrity failures
  • Collision avoidance recommendations conflicting with COLREGS
  • GPS/GNSS failures in high-risk areas
  • Software bugs causing erroneous navigation data

The “Two-Person Rule” Challenge
#

Maritime safety has traditionally relied on cross-checking between bridge officers. As AI takes over navigation functions:

  • Who cross-checks the AI?
  • How can officers verify AI recommendations in real-time?
  • What training prepares officers to identify AI errors?
  • When should officers override AI recommendations?
Automation Complacency
Maritime incident investigations have repeatedly identified “automation complacency” as a contributing factor, officers trusting automated systems rather than maintaining independent verification. The grounding of USS Guardian (2013) and numerous commercial vessels has been attributed partly to overreliance on electronic navigation without adequate chart verification. As AI systems become more sophisticated, this risk increases.

Maritime Liability Framework
#

Traditional Maritime Liability Principles
#

Maritime law includes unique liability doctrines that AI complicates:

Limitation of Liability:

  • Shipowners can limit liability to vessel value plus freight
  • 1976 LLMC Convention sets international limits
  • Question: Does AI failure constitute “actual fault or privity” defeating limitation?

In Rem Jurisdiction:

  • Ships themselves can be arrested and sued
  • Autonomous ships with no crew: who receives service of process?
  • How do traditional maritime liens apply to AI-caused damage?

General Average:

  • Cargo owners share extraordinary sacrifice for common benefit
  • If AI makes sacrifice decision, is it voluntary?
  • Can AI “reasonably” determine jettison necessity?

Autonomous Vessel Liability Questions
#

MASS presents unprecedented liability issues:

QuestionTraditional RuleMASS Uncertainty
Who is “master”?Ship’s captainRemote operator? AI? No one?
What is “navigational fault”?Master/crew errorAlgorithm defect?
When is vessel “unseaworthy”?Physical/crew deficiencySoftware bugs?
What is “privity” for limitation?Owner knowledgeAI decision opacity
What constitutes “negligence”?Human error standardAI behavior assessment

Allocation Among Multiple Parties
#

Autonomous vessel incidents may involve liability for:

  • Flag state (for certification and oversight)
  • Shipowner (for vessel condition and operation)
  • Operator (for navigation decisions, including remote)
  • Technology vendor (for AI system defects)
  • Classification society (for approval of autonomous systems)
  • Port/VTS (for traffic management failures)

International Regulatory Frameworks
#

United States Coast Guard
#

The U.S. Coast Guard has been developing policy for autonomous vessels:

Current Position:

  • Case-by-case review of autonomous vessel operations in U.S. waters
  • Requires determination of equivalent safety level
  • Existing manning requirements not waived without specific approval
  • Navigation rules apply regardless of automation level

Regulatory Developments:

  • 2022: Request for public comments on MASS
  • 2023: Continued participation in IMO MASS Code development
  • Ongoing: Review of domestic regulations for MASS compatibility

European Union
#

The EU has been active in autonomous shipping regulation:

Key Initiatives:

  • AUTOSHIP project: Demonstrating autonomous shipping technology
  • MUNIN project: Unmanned ship research
  • EU Maritime Safety Agency (EMSA) guidance on MASS
  • Support for Northern European MASS operational trials

Flag State Approaches
#

Different flag states are taking varied approaches:

  • Norway: Most permissive; operational autonomous vessels
  • UK: Maritime Autonomous Systems Regulatory Working Group
  • Singapore: Autonomous vessel testbed established
  • Japan: Major autonomous shipping development program
  • China: Extensive autonomous port and vessel development

Classification Society Standards
#

Role of Class Societies
#

Classification societies play critical roles in MASS safety:

Traditional Functions:

  • Independent verification of ship safety
  • Surveys and certification throughout vessel life
  • Rules for construction and equipment
  • Interface between flag states and industry

MASS Certification Challenges:

  • How to certify AI decision-making capability?
  • What testing validates autonomous navigation?
  • How to survey software systems?
  • Ongoing certification of learning systems

Classification Society MASS Rules
#

Major classification societies have developed autonomous vessel guidance:

Lloyd’s Register:

  • ShipRight procedure for autonomous ships
  • Design and operational requirements
  • Cyber security requirements
  • Remote operations guidance

DNV:

  • Recommended practice for autonomous and remotely operated ships
  • Technology qualification for MASS
  • Cyber security type approval

Bureau Veritas:

  • Guidelines for autonomous shipping
  • Notation for autonomous ships
  • Remote operations framework

Cybersecurity and Maritime AI
#

Maritime Cyber Threats
#

Maritime AI systems face significant cyber risks:

Documented Incidents:

  • GPS spoofing affecting ships in Black Sea region
  • ECDIS malware infections
  • Port system ransomware attacks (Maersk NotPetya, 2017)
  • AIS manipulation attempts

Autonomous Vessel Vulnerabilities:

  • Remote control channel compromise
  • Sensor data manipulation
  • AI decision-making interference
  • Supply chain attacks on software updates
GPS Spoofing Attacks
GPS spoofing, broadcasting false position signals, has become a significant maritime concern, particularly in regions of geopolitical tension. Ships have been shown phantom positions in the Black Sea and Persian Gulf, with AI navigation systems potentially misled. For autonomous vessels relying entirely on electronic position data, spoofing attacks could cause collisions or groundings. The standard of care increasingly requires verification systems and spoofing detection.

IMO Cybersecurity Requirements
#

IMO has established mandatory cybersecurity requirements:

  • Resolution MSC.428(98): Cyber risk management in safety management systems
  • Requirements effective January 2021
  • Flag states must verify cyber security arrangements
  • Applies to ships subject to ISM Code

Liability Case Studies and Precedents
#

Collision of MV ACX Crystal and USS Fitzgerald (2017)
#

While not involving autonomous systems, this collision illustrates navigation AI liability issues:

  • Container ship struck U.S. Navy destroyer
  • Seven Navy sailors killed
  • Investigations found failures in watch standing and navigation
  • AIS and radar systems were functioning but not properly monitored
  • Demonstrates consequences of inadequate human-machine interface

Implications for AI:

  • Navigation AI systems require proper human oversight
  • Training must address automation complacency
  • Liability extends to inadequate monitoring of automated systems

Ever Given Suez Canal Grounding (2021)
#

The Suez Canal blockage raised AI-adjacent liability questions:

  • 400-meter container ship grounded blocking canal
  • $9.6 billion in trade delayed daily
  • Complex liability involving owner, charterer, canal authority, insurers
  • Navigation AI and wind prediction systems involved

Standard of Care Questions:

  • Did AI weather systems adequately warn of conditions?
  • Were automated navigation aids properly functioning?
  • What role did algorithmic ship scheduling play?

Costa Concordia (2012)
#

The Costa Concordia disaster established precedents relevant to AI:

  • Captain’s override of automated systems
  • Questions about navigation system warnings
  • Criminal liability for captain’s decisions
  • Corporate liability for safety culture

AI Implications:

  • When should officers override AI recommendations?
  • What documentation of AI warnings is required?
  • How do we establish AI “said” something versus human decision?

Insurance and Maritime AI
#

Protection and Indemnity (P&I) Clubs
#

P&I clubs provide mutual insurance for shipowner liability:

MASS Coverage Questions:

  • Do existing P&I rules cover autonomous operations?
  • How are premiums calculated without crew history?
  • What cyber-related exclusions apply?
  • How is liability allocated for remote-operated vessels?

Hull and Machinery Insurance
#

Traditional hull insurance must adapt:

  • Coverage for AI system failures
  • Cyber attack damage coverage
  • Software defect versus physical damage distinction
  • Autonomous operation exclusions

The Nordic Plan and MASS
#

The Nordic Marine Insurance Plan (NMIP) has addressed autonomous vessels:

  • Provisions for unmanned vessels
  • Definition of “seaworthiness” for MASS
  • Requirements for shore-based operations
  • Cyber coverage clarifications

Establishing Maritime AI Standard of Care
#

Elements of Reasonable Care
#

The emerging standard of care for maritime AI includes:

Vessel Design and Equipment:

  • AI systems certified by recognized classification society
  • Redundant navigation and control systems
  • Cybersecurity meeting IMO requirements
  • Fail-safe mechanisms for AI failures

Operations:

  • Trained and certified remote operators
  • Adequate shore-based monitoring capabilities
  • Clear handoff procedures between automated and manual control
  • COLREGS-compliant collision avoidance logic

Training:

  • Officer training on AI system capabilities and limitations
  • Simulation training for autonomous vessel interaction
  • Cyber awareness training
  • Emergency procedures for AI failures

Documentation:

  • Complete voyage data recording including AI decisions
  • Maintenance records for AI systems
  • Software update history
  • Incident and near-miss reporting

Port Operations Standard of Care
#

For automated ports, the standard includes:

  • Segregation of automated and manned areas
  • Robust human-machine interface design
  • Emergency stop capabilities
  • Worker training on automated equipment interaction
  • Regular safety audits of automated systems

Frequently Asked Questions
#

What laws apply to autonomous ships in international waters?

Autonomous vessels in international waters are subject to flag state law and international conventions including SOLAS, MARPOL, and COLREGS. The IMO MASS Code, expected to be implemented by 2028, will establish international standards for autonomous vessel safety. Until then, flag states determine what autonomous operations they will permit for ships flying their flag. Different flag states have taken varied approaches, from Norway’s permissive stance to more restrictive positions elsewhere.

Can AI comply with COLREGS collision avoidance rules?

This is one of the fundamental challenges for autonomous vessels. COLREGS were written assuming human judgment, requiring action consistent with “good seamanship” and “ordinary practice of seamen.” Whether AI can exercise such judgment is debated. Current AI collision avoidance systems can follow specific rules but may struggle with ambiguous situations requiring seamanship judgment. The IMO is developing interpretive guidance, but many experts believe COLREGS will need amendment for fully autonomous operation.

Who is liable if an autonomous ship causes a collision?

Liability depends on the cause and may extend to multiple parties: the shipowner (for vessel operation), the remote operator (for navigation decisions), the AI technology vendor (for system defects), and potentially the flag state (for inadequate oversight) or classification society (for certification failures). Maritime limitation of liability conventions may apply, though AI failure might constitute “actual fault” defeating limitation. The allocation of liability among these parties is largely untested.

How are ports ensuring safety of automated operations?

Ports implement safety through: physical and electronic segregation of automated and manned areas; robust collision avoidance on automated equipment; emergency stop systems accessible throughout terminals; extensive worker training; regular safety audits; and incident reporting and investigation systems. International standards from organizations like PIANC and ICHCA provide guidance, though port automation safety requirements vary by jurisdiction.

What cybersecurity requirements apply to maritime AI?

IMO Resolution MSC.428(98) requires ships subject to the ISM Code to incorporate cyber risk management into their safety management systems, effective January 2021. This includes assessing cyber risks, implementing protective measures, detecting cyber incidents, responding to attacks, and recovering operations. For autonomous vessels, additional requirements may apply depending on flag state. Classification societies also have cyber security requirements for MASS notation.

What training is required for remote vessel operators?

The IMO MASS Code will establish competency requirements for remote operators equivalent to shipboard officers. Currently, requirements vary by flag state. Generally, remote operators should hold appropriate certificates of competency (deck or engineering as appropriate), receive specific training on remote operation systems, demonstrate competency in emergency procedures, and maintain ongoing proficiency. Several maritime training institutions now offer remote operator certification programs.

Related Resources#

On This Site
#

External Resources
#


Facing Maritime AI Liability Questions?

From autonomous vessel operations to port automation incidents to navigation system failures, maritime AI presents unprecedented liability challenges across international boundaries. With IMO MASS regulations approaching and autonomous shipping expanding, vessel operators, technology vendors, and ports need expert guidance on regulatory compliance, liability allocation, and insurance coverage. Connect with professionals who understand the intersection of maritime law, autonomous systems, and evolving international standards.

Get Expert Guidance

Related

Accounting & Auditing AI Standard of Care

The accounting profession stands at a transformative moment. AI systems now analyze millions of transactions for audit evidence, prepare tax returns, detect fraud patterns, and generate financial reports. These tools promise unprecedented efficiency and insight, but they also challenge fundamental professional standards. When an AI misses a material misstatement, does the auditor’s professional judgment excuse liability? When AI-prepared tax returns contain errors, who bears responsibility?

Advertising & Marketing AI Standard of Care

Artificial intelligence has transformed advertising from an art into a science, and a potential legal minefield. AI systems now write ad copy, generate images, target consumers with unprecedented precision, and even create synthetic spokespersons that never existed. This power comes with significant legal risk: the FTC has made clear that AI-generated deception is still deception, and traditional advertising law applies with full force to automated campaigns.

Architecture & Engineering AI Standard of Care

Architecture and engineering stand at the frontier of AI transformation. Generative design algorithms now propose thousands of structural options in minutes. Machine learning analyzes stress patterns that would take human engineers weeks to evaluate. Building information modeling systems automate coordination between disciplines. AI code compliance tools promise to catch violations before construction begins.

Childcare & Early Education AI Standard of Care

Artificial intelligence has entered the world of childcare and early education, promising to enhance child safety, support developmental assessment, and improve educational outcomes. AI-powered cameras now monitor sleeping infants for signs of distress. Algorithms assess toddlers’ developmental milestones and flag potential delays. Learning platforms adapt to young children’s emerging skills and interests.

Energy & Utilities AI Standard of Care

Energy and utilities represent perhaps the highest-stakes environment for AI deployment. When AI manages electrical grids serving millions of people, controls natural gas pipelines, or coordinates renewable energy integration, failures can cascade into widespread blackouts, safety incidents, and enormous economic damage. The 2021 Texas grid crisis, while not primarily AI-driven, demonstrated the catastrophic consequences of energy system failures.

Event Planning & Entertainment AI Standard of Care

The event planning and entertainment industry has embraced AI for everything from ticket pricing to crowd safety, but when algorithms fail, the consequences can be catastrophic. A crowd crush at a concert. Discriminatory ticket pricing. Facial recognition that wrongly ejects paying attendees. The standard of care for event AI is rapidly evolving as courts, regulators, and the industry itself grapple with unprecedented questions.