Marine surveying has long been regarded as a profession which has been shaped (not merely) by technical knowledge, but by experience, instinct and judgement, developed over years at sea, in shipyards and “on job training”. Traditionally, a marine surveyor arrives on board equipped with a notebook, camera, torchlight and (most importantly) a trained eye. Observations are handwritten, photographs are selectively taken and conclusions are drawn based on evaluation of accumulated knowledge of vessel’s operations, machinery behaviour and assessment of real-world failure patterns.
Today, however, the emergence of artificial intelligence has introduced a new dimension to surveying. An “AI Surveyor” can process documents, interpret images, analyse certificates and identify inconsistencies within seconds. Tasks that once required years of experience can now be (partially) replicated by a digital model.
Yet, the key question remains: Can AI replace the traditional surveyor, or does the future lie in a hybrid approach?
Defect Finding V Fact Reporting
In the field of marine surveying, no two attendances are ever identical. Whether assessing the aftermath of a casualty, evaluating an asset for pre-purchase, or tracing the origin of cargo damage, the job goes far beyond a standard protocol or a checklist. In this dynamic environment, the true measure of a surveyor’s efficiency hinges on his/her practical foresight and his/her ability to blend sharp visual observations (coupled with a mariner’s understanding of daily vessel operations). It requires an intuitive grasp of how machinery actually “behaves and fails” during its operation. The surveyor’s conclusion is anchored by decisive situational judgment.
A seasoned surveyor does not merely “find defects” but he/she interprets them.
For example, a newly built vessel leaving a shipyard may appear flawless to the untrained eye. However, an experienced surveyor may identify subtle misalignments like early stage coating failures, or installation inconsistencies that could later evolve into significant operational risks.
At its core, a valuable survey goes far beyond simply handing the master a punch list of defects (i.e a mere exercise of listing every defect). The challenge lies in rigorous application of risk prioritization. An expert surveyor acts as a critical filter, assessing which structural or mechanical anomalies pose an imminent threat to the vessel, her crew, the environment and which represent immediate operational vulnerabilities. They must accurately determine what constitutes a routine maintenance item suitable for onboard management, versus a material risk / onboard or shore management failure, requiring urgent escalation to H&M underwriters or P&I Clubs. This nuanced calibration of exposure and liability cannot be explained in a classroom; it is the culmination of years of frontline exposure, technical post- mortems and continuous hard-earned experience.
Artificial intelligence introduces a fundamentally different capability. AI based systems can:
- Analyse large volumes of documents instantly
- Interpret handwritten notes (with OCR integration)
- Review certificates for validity, expiry, and compliance
- Identify inconsistencies in reports and documentation
- Cross-reference data across multiple sources
- Generate structured, insurer-grade reports
A “fresh surveyor” equipped with AI tools can, in theory, perform at a level that previously required years of experience. For instance, AI can:
- Flag expired statutory certificates
- Detect missing documentation
- Identify inconsistencies between reports and supporting evidence
- Analyse photographic evidence for visible defects
- Compare maintenance records with expected standards
This significantly enhances efficiency and reduces the risk of human oversight. However, AI operates within a defined framework only. It identifies what is visible and structured, not necessarily what is operationally significant. The fundamental difference between a traditional surveyor and an AI-driven surveyor lies in practical interpretation.
AI can identify that:
- A certificate is expired
- A report is incomplete
- A parameter is outside standard limits
But it cannot fully understand:
- Whether a defect is operationally critical
- Whether a deviation is acceptable under real-world conditions
- Whether corrective action has already been effectively implemented onboard
Example: Consider a lubrication oil analysis report. An AI model may flag abnormal parameters and classify the condition as a defect. However, an experienced surveyor, after reviewing:
- Engine room logbooks
- Maintenance records
- Purifier operation logs
- Oil replacement history
May conclude that:
- The oil was recently purified after the report was issued
- Contamination levels have improved
- The machinery is currently operating within acceptable limits, as overhauls had just been completed
Thus, what appears as a “defect” on paper may not be a defect in practice.
This distinction is critical in marine surveying, where decisions directly impact:
- Vessel safety
- Insurance claims
- Operational continuity
- Financial liabilities
Modern shipping relies heavily on documentation. A vessel may appear fully compliant (on paper) with:
- Valid certificates
- Complete maintenance records
- Proper documentation
Yet, operational reality may tell a different story. A spotless document trail does not guarantee proper maintenance
or correct operational practices or actual equipment condition. The traditional surveyor bridges this gap by physically verifying the equipment’s operational condition, observes the crew practices and understands the overall behaviour onboard. The human interaction to understand the competency of the crew / officers.
Artificial intelligence is primarily document driven and has a fundamental limitation, when dealing with the physical realities of ships.
An inexperienced surveyor (relying on AI generative tools) might easily produce a voluminous, seemingly comprehensive report, but, if the physical investigation misses the mark, that massive document effectively buries the real issues.
The maritime industry operates at a relentless pace, such that vessels depart ports, cargo is discharged, repairs commence rapidly and volatile electronic data is quickly overwritten. True surveying capability means knowing exactly what to look at, and what to secure, in those critical first hours. Added to that is the responsibility of summarising the findings and documenting them in a “fair and logical” manner.
Ultimately, an AI-generated word count does not equate to a competent survey. A satisfactory report must strictly adhere to the instructing Principal’s mandate, present unvarnished facts and exclude any speculative tangents that could prejudice the instruction.
AI can validate documentation, but it cannot comprehensively validate reality.
Information Security: The Hidden Risk of the AI Surveyor
As the industry moves toward AI-driven surveying, a critical concern emerges regarding “data security”.
Marine surveyors routinely handle highly sensitive information, including:
- Insurance policies
- Claim documents
- Cargo details
- Vessel specifications
- Crew lists (personal data)
- Commercial agreements
- Financial records and invoices
Uploading such data into Ai systems, raises serious questions:
- Where is this data stored?
- Who has access to it?
- Is it being used for model training?
- Can it be leaked or misused?
Responsibility of a surveyor extends beyond technical assessment, it includes “data stewardship”.
International standards provide guidance on handling sensitive information, such as:
- ISO/IEC 27001 (Information Security Management Systems)
- Data protection regulations (GDPR and similar frameworks
Surveyors and organizations must ensure:
- Secure data storage
- Controlled access
- Encryption of sensitive files
- Clear data usage policies
Failure to manage this properly can lead to:
- Breach of client confidentiality
- Legal liabilities
- Loss of professional credibility
Legal and ethical responsibilities of surveyors
Unlike generic data, survey information is often owned by the client, not the surveyor.
This includes:
- Insurance data
- Vessel operational details
- Crew personal information
- Commercially sensitive records
The surveyor acts as a custodian, not as an owner.
When integrating AI into surveying workflows, the following must be considered:
- Has client consent been obtained before using AI tools?
- Are third-party platforms compliant with data protection laws?
- Is the data anonymised where necessary?
- Are there clear contractual terms regarding data usage?
The legal implications of mishandling such data can be severe, particularly in international maritime operations
Rise of the Hybrid Surveyor
The future of marine surveying does not lie in choosing between AI and traditional methods, it lies in integration. An effective surveyor of the future will be a Hybrid Surveyor, combining:
From Traditional Surveying:
- Experience
- Practical understanding
- Operational judgement
- Risk-based thinking
From AI Systems:
- Speed and efficiency
- Data analysis capabilities
- Documentation review
- Structured reporting
In practice, this means:
- Using AI to pre-analyse documents before attending a survey
- Leveraging AI to identify potential red flags
- Validating AI findings through physical inspection
- Using experience to interpret AI-generated insights
The above approach (of Hybrid surveyors) enhances both “Accuracy and Productivity”, while maintaining professional integrity.
Training the next generation
One of the most significant impacts of AI, will be on training of new surveyors.
Traditionally, surveyors developed their skills through:
- On-site experience
- Mentorship
- Progressive exposure to complex cases
With AI tools, new surveyors may:
- Rely heavily on digital outputs
- Skip foundational learning processes
- Develop limited practical understanding
This creates a risk of creating “Technically capable, but operationally inexperienced marine surveyors”.
To address this challenge, training programs must evolve to:
- Emphasise practical exposure
- Teach critical thinking beyond AI outputs
- Develop judgement and decision-making skills
- Encourage questioning of AI-generated conclusions
AI should be treated as a support tool, not as a substitute for learning
Conclusion: A question of balance
It is difficult and perhaps unnecessary to conclude which is superior: the “AI Surveyor” or the “traditional Surveyor”. The real value lies in balance.
AI can enhance efficiency, reduce oversight, and support decision-making. However, it cannot replace the human ability to:
- Interpret context
- Understand operational realities
- Exercise judgement under uncertainty
Marine surveying is not merely about identifying defects and submitting lengthy reports, it is about understanding what matters, what does not, and why.
In an industry where safety, liability and trust are paramount, the role of the surveyor remains fundamentally human.
The rapid acceleration of maritime technology has severely outpaced the evolution of statutory frameworks and case law, frequently resulting in jurisdictional mismatches
and protracted claims disputes. By the time complex technological disagreements reach apex tribunals – such as the Supreme Court or the House of Lords – the technological landscape has often already shifted, rendering newly established legal precedents practically obsolete upon publication. To mitigate this, regulatory bodies must pivot from reactive legislation to anticipatory governance.
Furthermore, the integration of Artificial Intelligence introduces a critical vulnerability regarding trust. Marine insurance is foundationally anchored in the principles of indemnity and Uberrimae Fidei (utmost good faith). As surveyors increasingly rely on modern technology, underwriters face two paramount concerns:
- First, can the AI-assisted findings of a surveyor be validated as strictly accurate?
- Second, there is a severe risk of proprietary vessel, voyage, or cargo data inadvertently entering the public domain through open-source AI models, thereby exposing the Assured, the Insurer, or the Charterer to unforeseen liabilities.
Challenge lies in safeguarding confidential information and maintaining good faith.
It is my considered opinion that the Insurers must mandate a disclosure clause, such that whenever a marine surveyor submits a report, they must specifically declare the type, scope and specific AI tools utilized in their assessment.
The future is not about replacing the surveyor with AI, It is about empowering the surveyor with intelligence, both human and artificial.
By Joel Lloyd Pinheiro FIIMS Marine Surveyor and Consultant
Quest Marine LLC (ISO 9001, 14001, 45001 ABS Certified Company)