Since ancient times, fishermen Indonesia have had to rely on their experience and intuition to find places with lots of fish. However, they could never be sure of their daily catch.
That all changed two years ago when officials from the National Research and Innovation Agency (BRIN) introduced an AI-based mobile app designed to help remove uncertainty in fishing.
The app, called NN Marlin, uses machine learning to process satellite and remote sensing data to identify fishing grounds with large fish stocks based on sea surface temperature and chlorophyll levels.
While other wealthy countries rely on startups and large tech companies to lead AI development, BRIN has partnered with local and foreign technology and research institutions to develop AI applications tailored to Indonesia's specific needs.
To date, BRIN has developed AI applications to study climate change, predict forest fires, monitor mangrove deforestation, and improve disaster management. While investment and adoption of AI in Indonesia still lags behind neighboring countries like Malaysia and Singapore, BRIN is working to close the gap.
In addition, BRIN also helps strengthen national research and innovation capabilities by ensuring that AI development involves local scientists and researchers, thereby maintaining national interests.
According to a study by consulting firm Kearney, AI could contribute nearly $1 trillion to Southeast Asia's GDP by 2030 and boost Indonesia's GDP by about 12%.
AI is seen as key to Indonesia achieving its goal of becoming a high-income economy by 2045 and achieving “more inclusive and sustainable economic growth,” said Teguh Prayogo, a researcher at BRIN.
In addition to assisting fishermen, NN Marlin is also used to detect illegal fishing activities and track foreign fishing vessels.
Indonesia, the world's largest archipelago nation, loses at least $3 billion a year to illegal fishing, according to government estimates. Teguh said BRIN is exploring ways to use AI to better detect illegal and suspicious activities.
BRIN also uses machine learning to assist law enforcement agencies. Recently, North Sumatra police used remote sensing technology in an application developed by BRIN to locate illegal cannabis fields. The technology can distinguish cannabis plants even in the middle of a forest by the quality of light reflected from the leaves.