On July 10, 2026, Konsemi was invited to attend the themed salon "Core Power: Empowering AI+AR Glasses." At the event, Liang Huaihua, General Manager of the Market Planning Department, delivered a keynote speech titled "Aspiring High with 'Core' Storage, Computing the Future with Intelligence – Storage Chips Redefining the New Boundaries of AI+AR Glasses."

As a representative Chinese enterprise with full-stack self-developed capabilities in the embedded storage chip sector, Konsemi showcased its comprehensive storage solutions for AI+AR smart glasses across all usage scenarios. The company demonstrated the strategic positioning and breakthrough pathways of domestic storage chips amid the rapidly evolving on-device AI landscape.

Starting from the fundamentals of storage architecture, the speech provided an in-depth analysis of the core tensions smart glasses face in balancing compute power, power consumption, physical footprint, and interactive performance. Konsemi offered a systematic response built around the triad of "small footprint, high performance, and ultra-low power consumption."

On-Device AI Deployment Demands a Storage Power Breakthrough

The accelerating deployment of on-device large models is propelling AI+AR glasses from being "minimalist information displays" to fully fledged "24/7 AI companions."

Emerging use cases such as multimodal interaction, real-time translation, and visual SLAM mapping are placing unprecedented demands on terminal data throughput and response latency. Ultimately, these performance requirements converge on the underlying storage chips—where the real bottlenecks lie.

Yet, as inherently compact wearables, smart glasses typically offer less than one cubic centimeter of internal space within the temple arms. Batteries, motherboards, optical modules, and sensors already consume the vast majority of that real estate, leaving only a sliver of room for storage components.

At the same time, the intensive read/write operations driven by multimodal AI tasks demand that storage chips deliver high bandwidth, robust error correction, and exceptionally low latency.

Achieving an optimal trade-off between capacity, performance, and power consumption within such extreme space constraints has become a critical hurdle to enhancing the smart glasses experience—and a formidable "storage power" challenge that chip designers must squarely address.

Small Footprint, Massive Impact

To tackle these industry pain points, Liang Huaihua highlighted Konsemi's small-footprint eMMC embedded storage products, purpose-built for compact devices such as AI glasses and smartwatches.

Key advantages of Konsemi's small-footprint eMMC include:

1. Ultra-compact packaging to conquer space limitations. Its highly miniaturized package design allows seamless integration into the restricted temple-arm space of smart glasses, adding no extra weight or thermal overhead.

2. Low-power engineering to extend battery life. With an intelligent power management unit supporting deep sleep and other low-power states, it significantly prolongs device endurance while mitigating heat buildup in the temple area under heavy loads.

3. Proven reliability to underpin on-device AI. Equipped with a fourth-generation LDPC error-correction algorithm, it ensures stable, high-throughput data handling for massive multimodal AI workloads, building a solid "storage foundation" for the smooth execution of on-device large models.

Liang Huaihua emphasized that Konsemi delivers a "storage-computing synergy" enablement model, architected from the ground up specifically for AI+AR glasses. Through innovative storage-chip design, Konsemi helps device manufacturers overcome the physical barriers of space and power consumption, enabling smart glasses to evolve from merely "functional" to genuinely "delightful" in user experience.

Full-Stack Self-Development Powers the Industry Forward

The "Core Power: Empowering AI+AR Glasses" salon was co-organized by the IICIE International Integrated Circuit Innovation Expo, the Shenzhen Augmented Reality Technology Application Association, and Core Master (Xinshiye). The event brought together leading players from on-device computing, chip design, storage, and other critical segments, aiming to foster deep collaboration across the industrial chain and chart the evolution from underlying hardware to real-world applications.

Through its active engagement and technical exchange at the salon, Konsemi gained clearer insights into the differentiated storage requirements of AI+AR glasses, while offering the industry a complete reference framework—from chip architecture and solution validation to final product deployment.

In her closing remarks, Liang Huaihua noted that as on-device AI penetration continues to rise, the demand for storage chips in smart glasses will increasingly converge on "larger capacity, smaller dimensions, lower power draw, and enhanced reliability."

Looking ahead, Konsemi remains committed to advancing its core embedded storage technologies. In close collaboration with ecosystem partners, the company aims to "aim high with storage and shape the future with intelligence," injecting sustained and robust "core power" into the widespread adoption of AI+AR glasses.