The Present and Future of MCUs and AI Semiconductors Opening the Era of On-Device Inference

 

Edge AI and AI Semiconductors: Present and Future

 

AI technology no longer remains confined to the cloud.

Within the practical constraints of power, latency, security, and cost, AI is increasingly moving into devices, and

at the center of this transformation are the evolution of AI semiconductor architecture and MCU-based edge AI technology.

 

In this webinar, we explore the overall technology trends and market changes in AI semiconductors,

and examine how edge AI inference is being implemented and commercialized in actual MCU environments as a coherent flow.

 

---

💡 Webinar Key Topics

(1) AI Semiconductor Technology Trends and the Rise of Inference-Centric Architecture

 

✔️ Changing AI semiconductor market structure following LLM proliferation

✔️ Separation strategies and technical background of training and inference

✔️ Architecture evolution centered on GPU, AI Accelerator, and HBM memory

✔️ AI system limitations from perspectives of computational performance, memory bandwidth, and power efficiency

✔️ Industry shift from cloud-centric AI to inference-specialized semiconductors

 

(2) MCU-Based Edge AI Practical Implementation Strategy – STM32 Case Study

 

✔️ Edge AI and on-device inference architecture based on STM32 MCU

✔️ Lightweight approach for running AI models with limited computational resources

✔️ AI applications based on sensing data

  – Utilization of vibration, current, acceleration, and voice data

✔️ Implementation cases of predictive maintenance and virtual sensor applications

✔️ Development flow from dataset to model generation to MCU deployment using

  STM32Cube.AI and NanoEdge AI Studio

 

---

👀 Recommended For

✔️ Embedded engineers reviewing edge AI and on-device AI implementation

✔️ Developers who want to understand AI semiconductor and inference architecture trends

✔️ Product and system designers looking to add AI capabilities to MCU-based systems

✔️ Technical practitioners considering AI implementations that meet low-power, real-time, and security requirements

✔️ Technical leaders seeking alternatives to complement the limitations of cloud AI

 

---

The evolution of AI semiconductors and practical implementation of edge AI,

new products and systems are born at the intersection of these two flows.

 

Through this webinar, consolidate your understanding of the big picture of AI semiconductors through practical application strategies for MCU-based edge AI all at once.