This week we had the pleasure of attending Oslo Ocean Days, bringing together people from the ocean industries to exchange ideas, challenge established ways of working, and discuss how technology can help shape the future of the maritime industry.
Today our founder and CEO, Kristin Omholt-Jensen, took part in a panel discussion exploring how we can turn the growing volumes of ocean data into actionable intelligence, and connect data, AI and digital technologies to make better decisions and build scalable operational capabilities.
The discussion raised some important questions for the maritime industry. Where can better data and AI materially improve decisions today? What is still preventing these technologies from delivering value at scale? How can maritime intelligence support commercial operations, compliance and voyage planning? And what will it take to build the trust, standards and data foundations needed to make AI-enabled decision support part of everyday workflows?
We have gathered some of the questions raised during the panel and asked Kristin to share her perspectives in more detail.
“What is one decision in ocean operations that can be materially improved today through better data or AI—and what is still preventing that improvement at scale?”
Without connectivity and access to clean, structured, and maritime-trained data, AI is of little value.
A digital infrastructure, enabled by affordable connectivity solutions such as Starlink, now makes it possible to connect vessels and shore-based operations in near real time.
Combined with structured, machine-readable data and AI models trained to understand the complexities of maritime markets and operations, organizations can automatically analyze information, identify patterns, and receive tailored alerts whenever conditions deviate from normal expectations.
This creates the foundation for a new generation of knowledge-based and data-driven decision support systems, capable of influencing and improving almost every operational, commercial, and strategic decision made across a maritime organization.
"...and what is still preventing to scale?"
Assuming connectivity is available, reliable, affordable, and predictable, the next challenge is building the right data foundation.
First, it is important to remember that not all data is equal. Data comes from many different sources: text, commercial data feeds such as raw AIS data, structured and unstructured datasets, proprietary formats stored in Excel and other systems, semi-open protocols, photos, videos, conversations, and information from the public internet. Bringing all of this together into a unified, trusted, and machine-readable foundation is a significant undertaking.
The next challenge is the maturity of AI technology itself. While AI has made remarkable progress, it is not a free resource. AI models and the infrastructure required to train and run them are both computationally intensive and energy demanding. AI is an incredibly powerful enabler, but its value depends entirely on having reliable connectivity and high-quality data that has been structured and trained to understand the context in which it operates. We are moving in the right direction, but we are not fully there yet.
Perhaps the most difficult barriers to overcome are not technological at all. Legal agreements governing data ownership, privacy, and sharing often limit how data can be used across organizations and ecosystems. Equally important is the human factor. Established ways of working, resistance to change, and a lack of trust in new technologies can slow adoption significantly.
In the end, scaling AI-driven decision support is not primarily a technology problem. It is a challenge that combines connectivity, data quality, AI maturity, legal frameworks, and, above all, people's willingness to embrace a new way of working.
“How can structured maritime data and AI be translated into concrete commercial decision support for the different companies involved in shipping—and where do you see the greatest impact today?”
Historically, data has primarily been used for analysis, and that analysis has largely been based on historical data. The tools of choice have been Excel, Power BI, and similar reporting platforms. More advanced users have applied Python, machine learning, and statistical models to uncover patterns and generate insights.
Today, however, you can have cathegorized, cleaned and defined historical data together with real-time data streams. This not only provides greater visibility into what is happening right now, but also enables near real-time predictions about what is likely to happen next. That fundamentally changes how decisions are made.
Another major shift is the ability to dramatically reduce the time spent searching for information. Most professionals know that the information they need exists somewhere within their organization, but finding it can be time-consuming and frustrating. Tools such as Microsoft Copilot are already helping people locate and utilize information more efficiently. However, companies like Microsoft focus on generic productivity use cases across all industries. Maritime Optima and other maritime technology providers can go much further by delivering AI solutions that understand the specific terminology, workflows, and challenges of the shipping industry.
Machines also excel at calculations. Today, they can evaluate thousands of scenarios simultaneously, compare alternatives in real time, and provide recommendations based on complex variables that no human could process manually. Humans remain responsible for making the final decisions, but increasingly, organizations may choose to let machines make certain operational decisions autonomously.
Finally, AI and automation have the potential to eliminate a significant amount of manual office work. Tasks such as preparing fixture notes, documenting decisions, creating handover reports, and updating systems can be automated or semi-automated. Office work has not experienced a true productivity revolution since managers stopped relying on secretaries and started using Word and email themselves. AI may well become the catalyst for the next major transformation of knowledge work.
The winning formula is access to trusted information, software that helps structure and interpret that information, and people with the knowledge, experience, curiosity, and critical thinking required to challenge the results and continuously experiment with new approaches
"How can maritime intelligence improve decisions related to compliance, voyage planning, routing, and commercial operations?"
I have already mentioned a few examples, but let's spend a moment on compliance.
What do we actually mean by compliance?
In 2026, compliance affects virtually every company connected to shipping, whether you are manufacturing equipment, chartering vessels, buying and selling ships, financing maritime assets, or providing insurance and professional services.
Organizations increasingly need to understand who owns a vessel, where it has traded, whether ownership has changed, whether the vessel has changed flag, MMSI, or name, and whether mandatory AIS transmissions have been interrupted or switched off. The combination of historical events and real-time intelligence provides a much deeper understanding of risk and behaviour. When relevant changes occur, automated alerts can notify you immediately, allowing you to ask the right questions before potential issues develop into serious problems.
Voyage planning is another area where maritime intelligence can have a significant impact.
Approximately 80% of voyage costs are driven by distance sailed and fuel consumption. Yet, unlike road transportation, there are no fixed roads at sea, apart from regulated Traffic Separation Schemes. Routing decisions are heavily influenced by weather conditions, commercial constraints, fuel prices, emissions regulations, port congestion, and bunkering opportunities.
Having access to historical voyage data in the proper operational context can therefore be extremely valuable. It is not enough to know which route a vessel followed; you also need to understand why that route was chosen.
Perhaps the vessel accelerated to meet a laycan window or a berth availability slot. Perhaps adverse weather was expected along the shortest route. Perhaps the operator wanted to bunker at a specific port with better fuel prices or more suitable fuel grades. Understanding the reasoning behind past decisions helps organizations make better decisions in the future.
In shipping, we often distinguish between pre-fixture and post-fixture activities. While this distinction is traditionally used for commercial execution and reporting, it also creates powerful opportunities for learning and continuous improvement. Historical AIS data, combined with navigational, weather, port, fuel, and commercial information, provides a rich source of intelligence that can help both shore-based teams and crews onboard optimize future voyages.
The real value emerges when historical knowledge, real-time data, and AI-powered analytics work together. This transforms maritime intelligence from a reporting tool into a decision-support system, enabling organizations to become more compliant, more efficient, and ultimately more competitive.
Do shipowners, charterers, operators, brokers and other maritime companies require different forms of decision support from the same underlying data?
I don't think anyone is being forced to adopt AI or data-driven technologies. However, shipping is a highly competitive and global industry, characterized by countless transactions and constant pressure to improve margins. Success has always depended on asking better questions, finding better information, and making better decisions.
Curiosity has always been a driving force in shipping. The industry's most successful people are often those who continuously seek new insights, challenge assumptions, and look for opportunities others have overlooked.
Today, most people still do not fully understand the potential of data and the wide range of AI technologies that are emerging. Yet they are curious enough to start exploring. They understand that there is value hidden in the vast amounts of data generated across vessels, ports, cargoes, markets, and organizations, even if they are not yet sure how to unlock it.
Those who begin experimenting now will learn faster, discover new opportunities sooner, and be better positioned to take advantage of the transformational changes that AI and data-driven decision-making are likely to bring to the maritime industry.
What does it take to make maritime data sufficiently structured, reliable, and accessible for AI-enabled decision support?
It takes far more than technology.
First, it requires a deep understanding of the shipping industry. Maritime operations are complex, and data only becomes valuable when it is interpreted within the context of commercial shipping, vessel operations, regulations, ports, cargoes, weather, and maritime workflows.
Second, it requires a highly skilled technology team. Building a maritime intelligence platform demands expertise across software engineering, data engineering, data science, user experience, and AI. Just as importantly, these professionals must be willing to learn the realities of shipping and work closely with maritime domain experts.
Third, it requires a way of working that brings together people with very different backgrounds and perspectives. Shipping experts, developers, data scientists, designers, and product managers must learn to speak a common language and collaborate around a shared objective. This interdisciplinary collaboration is often where the real innovation happens.
It also requires access to large volumes of data from multiple sources, combined with the patience and discipline to clean, validate, structure, and continuously improve that data. Building a trusted maritime data foundation is measured in years, not months.
Above all, it requires passion, persistence, and patient investors. Creating structured maritime intelligence that AI can understand and reason over is a long-term undertaking. There are few shortcuts. It takes vision, commitment, and the willingness to invest long before the full value becomes visible.
The result, however, is transformative: a maritime intelligence platform that can convert vast amounts of data into actionable insights, helping people make better decisions across compliance, commercial operations, voyage planning, and fleet management.
Where does AI create additional value beyond organising, visualising and making maritime data more accessible — and what determines whether users adopt the tool in their everyday workflow?
Organising, visualising, and making data accessible is only the starting point. The real value of AI emerges when it helps people make better decisions, faster and with greater confidence.
In shipping, professionals spend a significant amount of time searching for information, validating assumptions, comparing alternatives, and calculating scenarios. AI can dramatically accelerate these tasks. It can identify patterns hidden across large datasets, detect anomalies, predict future outcomes, recommend actions, and automatically generate insights that would otherwise require considerable manual effort.
AI also enables the combination of historical data, real-time operational data, market information, and company-specific knowledge. This creates a richer context for decision-making and allows organizations to move from reporting what happened yesterday to anticipating what is likely to happen tomorrow.
User adoption, however, is determined by more than technology.
Ultimately, people do not adopt AI because it is innovative. They adopt it because it helps them achieve better outcomes. In a highly competitive industry such as shipping, where margins are constantly under pressure and decisions must often be made quickly, any tool that helps people find opportunities, reduce risk, save time, or improve profitability will naturally become part of their daily workflow.
Who owns decision quality when the data, model, platform, and operator come from different organisations?
Ultimately, the responsibility lies with the decision-maker.
Data providers, technology companies, AI model developers, and platform vendors all have a responsibility to provide accurate, transparent, and reliable inputs. However, they do not make the final decision. The person or organisation acting on the information remains accountable for the outcome.
AI and data can improve decision quality, reduce uncertainty, and highlight risks and opportunities, but they should support human judgement rather than replace it. Just as a ship's master remains responsible for the vessel despite relying on charts, weather forecasts, and navigation systems, business leaders remain responsible for the decisions they make using digital tools.
What should be standardised or shared across the ecosystem, and what should remain proprietary?
In my opinion, data formats should be open and accessible to everyone.
Open standards create interoperability, which in turn drives competition and innovation. Without common standards, every company becomes trapped in its own ecosystem, making collaboration difficult and limiting the value that can be created from data.
I experienced this firsthand in the construction industry. The introduction of the open IFC standard forced software vendors, including large enterprise software providers, to support the export and exchange of data in a common format. This opened the door for an entirely new generation of BIM applications, each focusing on different parts of the value chain. The result was improved collaboration, lower costs, reduced waste, and fewer misunderstandings across projects.
Shipping could benefit from a similar approach. Open standards for data exchange would create a foundation upon which companies could build new solutions, services, and business models.
That does not mean everything should be shared.
While formats and interfaces should be open, the insights, workflows, customer relationships, analytics, and proprietary intelligence built on top of the data can remain competitive differentiators. In other words, the plumbing should be shared; the value created from it does not have to be.
Who has the right to standardise and define data?
That is a difficult question.
Large technology vendors often believe they should define standards because they have the scale and resources to drive adoption. Governments may argue that they should set standards to ensure transparency, security, and fair competition. Industry associations frequently position themselves as neutral parties capable of bringing stakeholders together.
Historically, industry-led organisations have often been successful. The IFC standard, for example, emerged from buildingSMART International and became a global standard for exchanging building information.
The challenge is that standards processes can move slowly. By the time committees agree on a framework, technology may already have moved on.
Perhaps the maritime industry needs both approaches: formal industry standards that create long-term stability and innovative companies that move faster, experiment, and demonstrate what is possible. Sometimes the fastest way to establish a standard is simply to build something that works and prove its value to the market.
Sharing data is where the commercial reality begins
Whether data should be shared is ultimately a business decision.
Companies will share data when they believe the benefits outweigh the risks. Trust, mutual value creation, and commercial incentives are therefore essential.
In practice, organisations are often willing to share data with partners, customers, suppliers, and service providers when doing so improves efficiency, reduces risk, generates revenue, or creates a competitive advantage.
The challenge is establishing trusted frameworks that allow data to be shared securely while protecting commercially sensitive information.
Closing remarks
Data, in its raw form, has very limited value.
An AIS transponder produces millions of position reports every day, but these are little more than digital signals until they are combined with context, domain knowledge, and a clear purpose. A stream of coordinates does not tell you why a vessel changed course, why it slowed down, or why it entered a particular port.
The real value emerges when data is transformed into information, information into knowledge, and knowledge into action.
Data becomes truly valuable when you know what you want to achieve with it.
For organisations that have a clear objective, the right expertise, and the ability to connect and contextualise multiple data sources, data may well be the most valuable asset they own. Not because it tells them what happened yesterday, but because it helps them make better decisions about what to do tomorrow.
