AI is quickly becoming an integral part of everyday life. Samsung Electronics’ “AI for All” vision aims to provide users with enriching AI experiences throughout their daily lives by setting a new standard for next-generation devices across mobile devices, TVs and AI-powered home appliances.
In-device AI is at the heart of Samsung’s AI innovations. This technology allows AI to operate independently within the device and does not require a server or cloud. Users benefit from fast AI response times without the need for a network connection, reducing concerns about personal data leaks.
On-device AI has now evolved beyond simple functions to handle requests requiring large amounts of data – such as generative AI – using only the device’s internal resources.
Samsung Newsroom explores how Samsung is leading the world in AI innovations on devices.
No Cloud Required: Transforming the On-Device AI Experience
Citing the need for speed and security, much of the tech industry has focused on running complex generative AI in an on-device environment. With long-term investments dating back over a decade, Samsung has already customized AI features for many of its devices and positioned itself at the forefront of providing on-device AI experiences for daily life.
Samsung’s Galaxy AI, for example, offers features built into the company’s latest mobile and tablet devices.1 Now supports a total of 20 languages,2 Galaxy AI enables communication without language barriers, including real-time conversation interpretation and translation of messages and web pages.
For TVs, Samsung leverages its expertise as a global market leader to train AI models. Since 2020, the company has continuously improved its TV-specific processor with an integrated NPU (neural processing unit). The NQ8 AI Gen3 processor in the Neo QLED 8K 2024 model has 512 neural networks for improved picture and sound quality. By analyzing and modifying pixels, images and sound sources, the processor delivers superior upscaling, smooth movements and clear dialogue for a smooth viewing experience in any situation.
The key to on-device AI is to minimize model size while maintaining performance. To achieve this, Samsung focused on efficient implementation while using specialized AI training data tailored to specific tasks.
▲ Sunmin Kim and Cheulhee Hahm, Master, Visual Display Business, Samsung Electronics explain AI on devices.
Smaller and Faster: Samsung’s AI Optimization Research
As AI technology advances and data processing requirements increase, model optimization becomes crucial. In other words, high-performance AI models must be able to operate efficiently within the constraints of devices’ resources – including their processors, memory and batteries – while maintaining levels of performance and reliability for everyday use. . Key technologies for this include model compression, hardware optimization and data processing acceleration.
Samsung has secured various technologies in the area of model compression to reduce the size of AI models. The company has successfully developed lightweight and fast AI models through techniques such as quantization, which improves response speed by simplifying algorithms and optimizing calculation processes; pruning, which removes non-essential elements from a larger model; and knowledge distillation, which transfers knowledge from large models to smaller models.
Samsung also continues to innovate in hardware optimization and data processing acceleration to run AI quickly and efficiently. For example, flash usage technology partitions large AI models and significantly reduces memory usage. Additionally, the company has developed patented technology that is expected to enable rapid inference on low-end devices without NPUs to expand AI applications to various products. Other research is underway into increasing inference speeds through speculative decoding, a technology that predicts the outcomes of AI models, and into improving computing power by running various hardware simultaneously.
▲ Knowledge pruning and distillation
End-to-end expertise: from internal development to implementation
With expertise spanning hardware, software, components and end products, Samsung is at the forefront of AI innovation on devices, from research and development to product implementation.
From a device perspective, Samsung’s market leadership in mobile devices, TVs, home appliances and more represents its excellence in hardware and software. Samsung integrates AI into its own devices, allowing the company to leverage both internal development and open collaboration with industry partners to create hardware-optimized AI models.
Samsung’s expertise also includes system software to link hardware and application software. For example, the Tizen OS and NPUs in Samsung TVs power on-device AI for a smarter and enhanced viewing experience. The company also provides developers with Vision AI and Language AI SDKs, as well as machine learning APIs to facilitate training and inference of AI models.
Samsung Research, the company’s advanced research and development center, is focused on improving competitiveness using the specialized expertise of its global network of laboratories – from the development of AI model structures and compression from models in the UK and US to creating AI acceleration solutions in India and Poland.
From a component perspective, Samsung has continually advanced the performance of NPUs as a leading semiconductor manufacturer and plans to continue working with its partners to develop new ways to use AI in devices.
▲ Samsung introduced a new AI TV experience at the Samsung Developer Conference 2024 in October.
With its on-device AI capabilities expected to expand to more products and services, Samsung is pioneering the revolutionary changes that will see AI technologies transform everyday life. In the second part of this series, Samsung Newsroom will explore the company’s innovations in personalized AI.
1 Galaxy AI supported devices include Galaxy Z Fold6, Flip6, Fold5, Flip5, Fold4, Flip4, S24 series, S23 series, S22 series, S24 FE, S23 FE, Tab S10 series, Tab S9 series, Tab S8 series , etc.
2 Languages supported by Galaxy AI include Korean, English, Arabic, Simplified Chinese, Dutch, French, German, Hindi, Indonesian, Italian, Japanese, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Thai, Turkish and Vietnamese.