Ship’s captain believes ChatGPT told him to change course

Ship Paper Chart Navigation
Share it now
from John Conrad (gCaptain) gCaptain has received a worrying report from an American captain alleging that a shipowner has used an AI bot, possibly ChatGPT, to suggest changes to the ship’s itinerary. Although the captain has chosen not to heed the recommendations, they are concerned that others may not exercise the same caution.

gCaptain was asked not to reveal the captain’s name or specific details about the incident that could be traced back to him. Given the limited size of the US flag fleet, this request has severely limited our ability to report on history, but we can provide an overview of this disturbing report.

During a recent ship’s voyage planning – a process in which navigation routes, safety measures and logistical considerations are meticulously assessed and the crew work to optimize efficiency while mitigating potential hazards – several minor safety concerns regarding the ship’s route and cargo were met this worried ship’s officers.

The captain composed a message stating the concerns and emailed it to the shipowner for guidance. Typically in these situations, a shipping inspector working for the shipowner will discuss any concerns with various experts (e.g. experienced masters, cargo planners, engineers) of the officer and provide risk mitigation advice to the ship’s master.

In this case, the captain received a quick reply. The captain told gCaptain that the first and last paragraphs of the message contained pleasantries and appear to have been written by a human, but the risk mitigation advice “sounds like it was written by an AI.” This message contained notices that would have necessitated a course change during the voyage.

At this point, it’s difficult to prove that the message was created by an AI, but it’s certainly in a similar style to the AI-generated one. As a former captain, I can attest that the advice given was mundane and unlikely to endanger the ship, but the point of International Safety Management procedures is that captains with more experience ashore – and other shore managers, who may be privy to Critical information not yet shared with the captain – can confirm that there is no imminent danger.

We asked ChatGPT-4 for their opinion on the matter with the following prompt. “Is it okay for chatgpt to advise on security-critical operations such as ship voyage planning?” The AI ​​responded with the following:

As an AI language model, ChatGPT is not designed or certified to provide advice on safety-critical operations such as ship voyage planning. While ChatGPT can generate text based on the input it receives, its understanding of complex, real-world situations and security considerations is limited.

Safety-critical operations such as ship voyage planning require the expertise and judgment of experienced professionals who can fully assess risks and contingencies. Relying on AI-generated advice for such operations could lead to potentially dangerous situations and unintended consequences. It is important that any advice or suggestion from AI systems is carefully evaluated and verified by qualified human experts before implementation, especially in safety-critical areas.
ChatGPT’s reply to Is it okay for chatgpt to advise on safety-critical operations such as ship voyage planning?

Yesterday ChatGPT announced version 4 with advanced analytical processing. With this new AI engine, it will be even more tempting for farmhands and ship officers—who are often overwhelmed with paperwork—to use AI to respond to concerns.

Conversely, AI has the potential to uncover dangers and risk mitigation opportunities in travel planning that humans may overlook. gCaptain has received reports through ChatGPT that help identify potential hazards that were not considered during the voyage planning process. Preventing the use of AI altogether – or just making it difficult to use with comments like “I don’t trust these robots” – could discourage ship captains from using it and finding a new solution that protects their crew. This could even provide additional employment opportunities in the future, as ChatGPT suggestions could be reviewed and edited by an experienced captain working with a “fast engineer” (a fast engineer is someone capable of asking AI the right types of questions ).

Regardless of whether the use of AI can be confirmed, in this incident it is crucial that organizations such as IMO, USCG and INTERTANKO discuss and provide guidance on the implications of using AI, because – whether they now take the advice of the AI or not – it is only a matter of time before a major shipwreck occurs and lives are lost.

Meanwhile, shipping companies could just be more transparent and inform the captain of all messages sent to a ship, including the help of AI. This notification can be as simple as “ChatGPT was used to fix the grammar in this email” to “here are some additional ideas about this problem caused by Bing AI”.


Share it now
%d bloggers like this: