Synopsis: In this Techstrong AI video, Mike Vizard talks to Michael Wu, president and general manager for North America at Phison Technology, about how NAND memory technologies can be used to optimize usage of graphical processor units (GPUs) to improve artificial intelligence (AI) applications.
In this Techstrong AI video, host Mike Vizard interviews Michael Wu, president and general manager of Phison Technology, to discuss making generative AI more affordable and accessible for small and medium-sized businesses. Wu highlights the current challenges businesses face with generative AI, such as the high costs of GPUs and infrastructure required to train large models. He explains how Phison Technology is addressing these challenges through innovations like SSD-based adaptive caching, which expands GPU memory and reduces the need for expensive hardware, enabling smaller businesses to harness the power of AI without the prohibitive costs typically associated with large language models.
Wu also discusses the future of AI infrastructure, emphasizing that companies need affordable, secure, and easy-to-use solutions to integrate AI effectively. He introduces Phison Technology’s upcoming “Adaptive Pro Suite,” a software solution designed to simplify AI integration for businesses, allowing them to train models using their proprietary data in a cost-efficient and secure manner. Wu believes that while the AI revolution has garnered significant attention, its long-term success will depend on providing businesses with scalable, practical tools to see a return on their investment in AI infrastructure.
