Ted Hirokawa
March 4, 2025 08:58
Atlantes, a new AI model developed by AI2, supports conservatives who protect natural resources from illegal activities by improving monitoring of world fishing and wildlife.
Innovative AI models are aimed at illegal fishing.
In a breakthrough development for preservation efforts, researchers introduced an open source AI model called Atlantes designed to monitor global navigation vessels. Developed by NVIDIA, this model, developed by Alen Institute for AI (AII) in Seattle, aims to fight illegal fishing by analyzing 5 billion GPS signals every day in about 600,000 marine vessels.
High accuracy and real -time warning
Atlantes boast about 80%impressive prediction accuracy when determining ship activities. It is integrated into AI2’s skylight maritime monitoring platform to warn the authorities within 15 minutes after detecting potential illegal fishing activities. This ability was proven when the Argentina Navy was a warning of a mining window, intercepting illegal fishing vessels and imposing fines.
Atlantes’ technology backbone
The AI model consisting of 4.7 million parameters is based on the NVIDIA H100 Tenser Core GPU and PyTorch. From January 2022 to June 2024, we process automatic identification system (AIS) data, which is essential for most ships, which have more than 15 million signals for marine experts to improve the model’s precision.
A wider impact on global fishing
According to the Financial Transparency Union, illegal, un reported and unpression (IUU) fishing is up to $ 23 billion in global economic costs, accounting for about 20%of the world’s fishing. In particular, the influenced African waters are the African waters where the community depends greatly on fishing for living and employment.
Expand the AI by preserving wildlife
AI2 plans to integrate with Earthranger, a platform that compils data from various sources to monitor wild animals by expanding the use of Atlantes other than offshore applications. This includes traces of elephants, rhinos and wild dogs to alleviate human actresses’ lifelong conflicts. This system is trained to predict elephant behavior, reducing the conflict between elephants and farmers.
Future prospects and preservation effects
The AI2 aims to actively solve human’s’s lifting collisions by utilizing a wide range of data sets of elephant movements. Jes Lefcourt, director of Earthranger, emphasized the potential to save the life of an elephant by predicting exercise and preventing conflicts with humans.
Infrastructure, which supports the classification of fishing boats, can be applied to predict elephant behavior and shows the diversity of AI in preservation. This innovative approach can greatly improve the protection of natural resources and biodiversity worldwide.
Image Source: Shutter Stock