Expanding the NPU workforce to 2,000 by 2030
| NPU, a processor optimized for deep learning operations
Expanding NPU adoption from mobile to battlefield and IoT

Samsung Electronics announced on the 18th that it is putting its life on the line to foster its Neural Processing Unit (NPU) business.

By 2030, Samsung Electronics plans to expand its NPU workforce by more than tenfold to 2,000 people and focus its entire company's capabilities on strengthening "next-generation NPU technology."

By developing its own NPU technology, Samsung Electronics is accelerating the achievement of its "Semiconductor Vision 2030," which aims to become the global leader in system semiconductors by 2030.

NPU is a processor optimized for deep learning algorithm operations.

Deep learning algorithms require parallel computing technology to process thousands of calculations simultaneously, and NPUs are considered a key technology for implementing AI because they can efficiently perform these large-scale parallel calculations.
Samsung Electronics has been conducting advanced research and product development in its System LSI Business Division and Advanced Technology Research Center to strengthen its NPU technology capabilities, and as its first result, it has developed the 'Exi', which has its own NPU embedded in a mobile SoC.North 9 (9820) was introduced last year.

This product implements 'On-Device AI', enabling mobile devices to perform artificial intelligence computations that were previously performed by exchanging data with cloud servers.

Samsung Electronics plans to expand NPU adoption across all IT sectors, from mobile devices to battlefields, data centers, and IoT.

Samsung Electronics plans to sequentially expand the application of NPUs starting with its flagship SoC products for mobile devices, and will also focus on developing automotive SoC products equipped with NPUs, such as in-vehicle infotainment (IVI) systems and advanced driver assistance systems (ADAS).

We also plan to expand the scope of its use by developing a deep learning-specific NPU that can enhance the big data processing performance of data centers and strengthen AI operations.

Samsung Electronics also plans to expand ongoing collaborations with global research institutions and domestic universities, and invest heavily in discovering key talent.

Last May, Samsung Electronics expanded and relocated its Montreal AI Lab, a research institute specializing in deep learning, to the Mila Institute in Canada, and is collaborating with researchers from the University of Montreal and McGill University, led by world-renowned scholar Professor Yoshua Bengio.

Since 2017, we have been conducting joint research on artificial intelligence with domestic universities through the Neural Processing Research Center (NPRC).
NF2.png" style="width: 600px; height: 400px;" />
“We will take the lead in the coming AI era by strengthening our NPU business, which is at the core of deep learning algorithms,” said Kang In-yeop, President of Samsung Electronics’ System LSI Business. “We will present innovative processors that are one step ahead through differentiated technologies, collaborations with global organizations, and recruitment of key talent.”

Samsung Electronics plans to further develop NPU technology and expand it into neuromorphic processor technology that enables information processing and recognition at the level of the human brain.