The predominant narrative surrounding Artificial Intelligence is comfortably simple: AI is just a tool. Much like a calculator or a compass, the assumption has been that the responsibility for its application—and any resulting fallout—rests squarely on the shoulders of the user. But as AI systems grow exponentially in complexity and influence, this archaic viewpoint is no longer adequate. It is time we change how we view AI, moving away from a model of absolute user responsibility toward one of shared liability.

Consider how society manages other ‘tools’ that possess the potential to cause significant harm. We do not allow unregulated, universally accessible markets for restricted medications or firearms. Instead, we implement stringent safety mechanisms. We restrict access, demand rigorous qualifications to operate them, and crucially, we assign and share liability with manufacturers if a product proves inherently dangerous. When the potential for harm is high, society takes extraordinary care to protect its citizens, particularly the vulnerable. Yet, with AI, we have inexplicably abandoned this standard.

The technology industry currently presents the public with a glaring contradiction. On one hand, we are reassured that AI is merely an assistant, a passive aid meant only to support human decision-making, keeping the human firmly ‘in the loop.’ On the other hand, AI is aggressively marketed as a revolutionary force capable of gathering unfathomable amounts of data, making intuitive leaps, and delivering insights far beyond mere human capacity. We cannot have it both ways. If a tool is sold on the promise of delivering highly complex, critical outputs, the entities manufacturing and training these models must bear responsibility when those outputs cause real-world harm.

This shift in perspective is extremely relevant to the maritime sector, which carries the responsibility of maintaining the majority of world trade. If an AI vessel routing tool miscalculates weather patterns, leading a vessel into hazardous conditions, or a bridge system ‘hallucinates’ navigational data, the liability currently completely lies with the crew. 

While one would argue that employers should empower their teams to challenge the AI’s output if they are uncomfortable with it, we know that power dynamics between ship and shore can make this challenging, as being perceived as ‘difficult’ could eliminate seafarers from lucrative routes or newer vessels. Furthermore,the crew member who deviates from the AI ‘guidance’ will be asked to justify their decision, which makes it obvious that this is less of a tool to support human decision making and more of an efficiency measure that has the approval of the employer.  Similar power dynamics are at play on shore where teams are being asked to use AI to make important decisions on everything from new contracts and vessel designs through to interpreting regulations. 

Currently, individual vendor contracts are meticulously drafted to push all liability onto the end-user—the ship’s captain or the operating company. This creates a dangerous void in accountability. Industry experts have repeatedly noted that the International Maritime Organization (IMO) cannot effectively regulate maritime AI because the technology evolves far too quickly for traditional bureaucratic frameworks to keep pace. However, there is a powerful alternative to slow-moving international legislation: market-driven self-regulation through shared liability.

If we establish a wider regulatory norm of shared responsibility, the market is likely to naturally correct itself. If the manufacturers of a vessel routing tool know they will share the financial and legal burden of a shipwreck caused by their software’s hallucination, they will radically alter their approach. Shared liability forces manufacturers to prioritise rigorous testing, safety guardrails, and reliability over the race to rush a half-baked product to market.

Shared liability should not be left to the mercy of individual contract negotiations; it must become a foundational principle of AI deployment. By holding manufacturers accountable for the immense power they claim their tools possess, we ensure that the technologies of tomorrow are built safely for the people who rely on them today. More importantly, it acts as a legal airgap between maritime’s cautious digital evolution and tech’s ‘move fast and break things’ approach.
——–

Words: Namrata Nadkarni, Intent Communications founder and CEO.  Photo by Mari Helin on Unsplash

Like this article? Click here to sign up for our free monthly ‘De-Mystifying Digitalisation and AI in Maritime’ newsletter and get the latest cross-sector expert insights, news, interviews, videos and case studies straight to your inbox. You can also buy a ticket to the Demystifying AI in Maritime 2026 London Summit here.