Science creates the tools, but policy dictates how they are used. Last week, I had the privilege of traveling to Brussels for the EU Ocean Days, selected as one of 15 young representatives to engage in a direct policy dialogue with the European Commissioner for Fisheries and Oceans, Costas Kadis.
As a computational biologist, I often spend my days looking at the micro-level. But sitting in the European Commission, the conversation was decidedly macro: How do we use Artificial Intelligence to govern 70% of our planet's surface?
The Ocean as a Computational Problem
My intervention focused on the European Digital Twin of the Ocean (DTO). I argued that this initiative cannot simply remain a visualization tool—it must become an operational engine for management.
A "Digital Twin" is a virtual replica of the ocean that mimics its physical, chemical, and biological behaviors in real-time. But a model is only as good as its predictive power. I discussed with Commissioner Kadis how AI algorithms are the missing link to turn this static data into dynamic governance.
We discussed concrete applications of AI in ocean management:
- Dynamic Routing: Using AI to re-route shipping lanes in real-time to avoid whale migration paths or noise-sensitive ecosystems.
- Precision Fisheries: Moving away from annual quotas to dynamic limits based on predictive spawning models, ensuring stocks recover faster.
- Hypoxia Prediction: Forecasting algal blooms and low-oxygen zones weeks in advance to protect coastal economies.
The insights gathered from our dialogue are feeding directly into the European Ocean Pact, set to be presented in June 2025. This is a rare opportunity where youth perspectives—and technical expertise on Digital Twins—are being baked into the legislative foundation of the EU’s Blue Economy.
Democratizing the "Digital Ocean"
However, technology is not neutral. A key part of my discussion centered on access. If the EU builds a sophisticated Digital Twin, who gets to query it?
I advocated for a system where these high-performance predictive tools are accessible not just to industrial fleets, but to local communities and artisanal fishers. An algorithm that optimizes a catch or predicts a storm should be a public good. This is the essence of the One Health approach: healthy oceans, sustainable economies, and equitable access to technology must go hand in hand.
Final Thoughts
Leaving Brussels, I feel more convinced than ever that the next generation of scientists cannot afford to stay in the lab. We need to be fluent in two languages: Python and Policy. Only by bridging this gap can we ensure that the Digital Twins we build actually help us heal the real world.
View Official Event Page