I recently joined the Same Brain Podcast to geek out on the latest AI trends with Justine and Jenna Ezarik. We had a great discussion about how AI is changing the way we use our devices and get things done. I got to show off some of the coolest new products that run generative AI on-device, and proved that artistic talent isn’t required to make something amazing with image generation.
Using AI in our Daily Lives
Like a lot of people, I’m using AI tools every day to maximize my productivity and unleash my creativity, at work and at home. I talked to Jenna and Justine about how I have used chatbots to invent new cocktail recipes, write birthday poems, and settle family debates on obscure topics.
It’s helpful that many devices now have generative AI tools built into the apps that people already use, keeping you “in the flow.” Chromebooks are a great example since Gemini is integrated into tools like Circle to Search, which makes it easy to get information about whatever is on your screen, and Smart Grouping, which helps manage and organize different tabs.
It’s also really impressive to see the AI-powered creative tools on today’s devices. In the video, you can watch me put my questionable artistic skills to the test, walking through the image generation features on the Samsung Galaxy Tab S11, which has a MediaTek Dimensity 9400+ chip with a powerful Neural Processing Unit (NPU).
Agentic AI is the Next Step
Agentic AI is even more proactive and personalized than generative AI. Agentic AI is designed to learn about your preferences and then combine that information with other data to give you personalized assistance.
On the podcast, I shared one example of how agentic AI will make your life easier. A chatbot would automatically book you a ride to the airport, keeping in mind your preferred rideshare service and how early you typically like to arrive ahead of flights. That might seem like a small example, but Justine made a great point about how these AI tools can help alleviate the mental load of scheduling and coordinating so many things, which really makes a difference.
Processing Information On-device
Generative and agentic AI are rooted in the cloud, due to the massive performance and scale of data centers. However, we’re now seeing more processing move from the cloud to the edge. One reason is that companies like MediaTek are designing powerful NPUs that can process LLMs with billions of parameters at the edge. At the same time, we’re starting to see more SLMs that are optimized for specific workloads and work really well at the edge. Processing information on-device reduces latency to improve the user experience, and enables the AI to better understand you personally while protecting your privacy.
And it’s not just smartphones, tablets, Chromebooks and PCs where we’re seeing a demand for edge computing. NVIDIA’s DGX Spark is an AI personal supercomputer that allows developers to prototype and fine-tune AI models with up to 200 billion parameters locally. DGX Spark is able to process these massive workloads thanks to its GB10 Grace Blackwell Superchip, co-designed by MediaTek and NVIDIA, that is both super powerful and power-efficient at the same time.
Follow Adam's newsletter to learn more about the latest Innovation Talk.
