NEC Corp. has begun developing an AI model capable of analyzing acoustic data to quickly and accurately assess conditions in underwater environments, the Japanese company announced Tuesday.

The initiative, which is based on work awarded to NEC by the Japanese government's Acquisition, Technology & Logistics Agency, aims to strengthen Japan's defense capabilities, but the technology is also expected to be useful in exploring for marine resources such as rare earth elements, according to the announcement.

Light and radio waves cannot travel very far through ocean water, so sound waves are a more effective tool when probing for marine life and resources. However, the sounds of waves and moving ships create background noise, and interpreting marine audio data has traditionally required experts with specialized knowledge and experience to spend significant amounts of time analyzing it.

The AI that NEC plans to develop will be trained on a massive amount of underwater sound data to enable it to rapidly and accurately assess underwater conditions. It will utilize technologies employed in systems such as OpenAI's ChatGPT generative AI model and NEC's own proprietary generative AI. NEC has been developing sonar technology for detecting submarines and other objects for over 90 years and will leverage this expertise for the project as well.

Potential practical uses for the new AI system include analyzing sound waves detected by sensors mounted on ships and aiding in the detection of submarines and seabed resources, as well as weather forecasting. NEC also said in its announcement that, in the future, the system may be used for highly precise earthquake prediction.

----

This article is from The Yomiuri Shimbun. Neither Dow Jones Newswires, MarketWatch, Barron's nor The Wall Street Journal were involved in the creation of this content.

YDN-M0000225735-1