Recently, I had the pleasure of joining the Same Brain Podcast alongside renowned content creators Justine and Jenna Ezarik. Our conversation revolved around the transformative impact of AI on how we interact with our devices and accomplish various tasks. I also showcased innovative products that leverage on-device generative AI, demonstrating that one doesn’t need exceptional artistic skills to achieve impressive results in image creation.
Integrating AI into Daily Life
Like many others, I rely on AI tools daily to enhance productivity and foster creativity at both work and home. During our discussion, I shared examples of how I’ve utilized chatbots to concoct unique cocktail recipes, craft birthday poems, and settle family disputes on obscure topics.
Today, numerous devices incorporate generative AI features directly into commonly used applications, facilitating a seamless user experience. For instance, Chromebooks now have Gemini integrated into tools like Circle to Search, which simplifies the process of finding information relevant to what’s displayed on your screen, and Smart Grouping, which aids in the organization of multiple tabs.
The capabilities of AI-driven creative tools are also evolving. In the podcast, I tested my artistic abilities using the image generation functionalities on the Samsung Galaxy Tab S11, powered by a MediaTek Dimensity 9400+ chip equipped with a neural processing unit.
The Emergence of Agentic AI
Agentic AI marks a significant progression beyond generative AI, offering a more proactive and personalized experience. It is designed to learn individual user preferences and leverage that data for tailored assistance.
During the podcast, I outlined a scenario where agentic AI could streamline routine tasks. For instance, a chatbot could automatically arrange a ride to the airport, considering your choice of rideshare service and your typical arrival time before flights. Although it appears minor, Justine pointed out that alleviating the mental burden of scheduling can substantially improve our daily lives.
On-Device Information Processing
While generative and agentic AI largely rely on cloud computing due to the extensive resources needed by data centers, there’s a growing trend of transferring processing capabilities closer to the user. This shift is driven in part by advancements in powerful neural processing units, which can execute large language models with billions of parameters directly on devices. Concurrently, smaller, optimized language models are gaining traction on edge devices, offering efficient performance for specific tasks.
On-device processing not only minimizes latency and enhances user experience but also allows AI systems to provide more personalized responses while safeguarding user privacy.
This trend isn’t confined to smartphones, tablets, Chromebooks, and PCs. For example, Nvidia’s DGX Spark serves as an AI personal supercomputer enabling developers to prototype and refine AI models with up to 200 billion parameters locally. It manages these demanding workloads through the GB10 Grace Blackwell Superchip, co-designed by MediaTek and Nvidia, striking a balance between performance and energy efficiency.
Adam King is the Vice President and General Manager of the Personal Devices Business Unit at MediaTek, a Taiwan-based semiconductor company that specializes in designing system-on-a-chip (SoC) solutions for a wide array of consumer electronics.
In conclusion, the advancements in AI technology continue to reshape our interaction with devices, enhancing productivity and creativity in unprecedented ways. As both generative and agentic AI evolve, they offer exciting possibilities for making our daily routines more efficient and personal.
