By Sameer Sharma, AVP of MediaTek's Internet of Things Business (Americas & Europe).png?width=150&height=150&name=Sameer%20Sharma_Round%20(1).png)
For years, the promise of the Internet of Things (IoT) has been to connect devices, generate data, and help organizations operate more efficiently. The challenge has been analyzing that data in real time and translating it into action. AI is changing that by enabling connected devices to understand their environments, generate insights, and instantly make decisions.
To support these capabilities, we’re seeing more AI processing move to the edge. On-Device AI offers faster response times and improved privacy by running applications at the edge. Additionally, edge AI processing reduces reliance on the cloud, which is especially important for environments with limited or unreliable connectivity. Meeting these computing demands requires powerful, efficient hardware and software designed for AI at the edge.
The foundation of Edge AI devices
The MediaTek Genio family is designed specifically for AI-powered IoT devices, combining multi-core CPUs, GPUs, and NPUs for real-time edge inference, vision processing, and multimedia workloads. Our Genio platforms optimize performance and battery life by dynamically switching between high-performance and energy-efficient cores. Our Genio portfolio supports Wi-Fi, Bluetooth, and GNSS connectivity as standard, with external modules enabling 4G, 5G, or 5G/RedCap cellular connectivity. Our IoT portfolio also includes modem-based platforms with integrated 4G or 5G modems.
For demanding edge AI applications such as robotics and drones, we developed MediaTek Genio Pro 5100, which enables powerful on-device generative AI and real-time multimodal processing at the edge. Integrating our powerful 8th generation NPU, Genio Pro 5100 has over 50 TOPS of gen AI acceleration. To meet the durability and longevity needs of industrial, healthcare, and embedded applications, Genio Pro 5100 supports an industrial temperature range and a 10-year product lifecycle.
Software also plays an important role in the IoT, allowing developers to deploy the latest AI applications. The MediaTek Genio Developer Center provides engineering resources to accelerate IoT development, and the MediaTek NeuroPilot SDK gives developers the tools and APIs needed to optimize, deploy, and manage AI models across a wide range of devices. By supporting a single codebase across multiple platforms, NeuroPilot helps reduce development time and eliminates the need to rewrite applications for different IoT deployments.
MediaTek works closely with ecosystem partners to further simplify AI development. We have integrated our NeuroPilot SDK with the NVIDIA TAO toolkit to streamline model training and deployment, helping developers bring AI-powered products to market quickly.
Vertical spotlight: AI in action across different industries
What’s fascinating about AI is how it can be tailored to so many use cases and industries to meet different requirements. Below is a look at how AI is being used across some key verticals in the IoT.
Industrial
With the rise of physical AI, we’re seeing greater demand for automation and autonomy in industrial environments. Robotics and drones are being used to collect data, inspect facilities, make decisions in real time, and perform tasks in factories and other settings where conditions may be hazardous. The benefits include improved productivity, increased operational speed, and enhanced worker safety. Additionally, industrial environments are using human-machine interface (HMI) displays with AI capabilities to streamline workflows and intelligently visualize operational insights.
Retail
AI is creating new opportunities for customer engagement and more efficient store operations. Self-service AI kiosks and AI-powered point-of-sale systems can help shoppers locate products, offer frictionless scanning, and generate in-store analytics that reveal useful operational insights. AI retail agents can also answer questions and assist customers virtually and on the sales floor.
Healthcare
AI helps physicians and care teams streamline workflows so they can spend more time with patients. For example, clinical note-taking devices can capture important details during appointments, allowing providers to stay focused on the conversation instead of typing notes. Bedside monitors and IoT sensors can also deliver real-time analytics, early alerts, and personalized insights to support proactive care and continuous monitoring, while reducing the burden on clinical staff. AI is also helping healthcare professionals and researchers process information more quickly and accurately through applications such as medication identification and sorting, as well as microscope image analysis.
Education
Across primary, secondary, and higher education, schools are deploying AI-capable smart displays, projectors, and whiteboards. AI helps these tools display relevant, contextually aware information in each classroom, and across campus, while also enabling more personalized, interactive learning for students.
Bringing AI everywhere
MediaTek’s portfolio is designed for different performance tiers, making our technology ideal for a wide variety of IoT applications and use cases. Additionally, our silicon and software are designed for scalability so it’s easy for companies to leverage our technology across products and market segments.
As AI continues to evolve, the devices closest to the real world will increasingly be where the most important decisions are made, and that is exactly where MediaTek is focused.
To learn more, our case studies offer examples of how MediaTek’s IoT portfolio can make the difference in real-world devices. Learn more about how we’re driving the next generation of IoT intelligence.
