At Intent Communications, we love a good debate, and this is exactly what we got from our judges for the first round of our AI comparison project for maritime information.
Our judges were answering the question: Which AI model best explained why the IMO’s Net Zero Framework wasn’t adopted?
Contenders to the throne were the free versions of five leading large language models, ChatGPT, Claude, Copilot, Gemini and Grok – and Gemini was crowned the winner (you can find the scores on our website)!
Our judges Namrata Nadkarni and Ed Parshotam have captured their discussion of the scoring for you in a video podcast format available on Spotify (link to the episode below), but if you prefer a written summary, here are some key takeaways:
- Gemini offered a well-structured answer that clearly explained the political pressure, economic concerns, and the broader journey of the Net Zero Framework, including what happens next.
- Copilot also performed strongly, particularly in capturing the political dynamics behind the adjournment that some other models missed.
- Several models struggled with nuance. For example, some responses downplayed the level of emotional difference from key Member States or failed to explain the timeline towards the October 2026 decision point.
- Sources were another issue. Many of the models didn’t reference the IMO itself or shipping organisations, and some relied heavily on single sources or even social media posts.
🎧 You can also hear Namrata and Ed discuss the results here: Comparing AI for Maritime: Why wasn’t the Net Zero Framework adopted at the recent IMO MEPC Extraordinary Session? – De-mystifying digitalisation and AI in Maritime – by Intent Communications | Podcast on Spotify