When computational density crosses the cost threshold and privacy compliance becomes a daily necessity, edge AI is at a historic inflection point, transitioning from a "technical concept" to "large-scale commercial deployment." Within this global industrial restructuring, the home environment is emerging as a prime application arena for edge AI due to its hard requirements for real-time response, privacy, and scenario coordination. However, the smart home industry has long been trapped in the "single-device intelligence" silo dilemma, lacking a system-level brain that can unify multiple protocols, cross brands, and manage all scenarios. During this critical window, whoever can construct an ecosystem moat with "cloud-edge-device" synergy will gain the upper hand in defining the next generation of home intelligence.
SDMC (00901), leveraging its unique operator channel barriers and first-mover advantage within the Google ecosystem, is leveraging its "1+2+X" strategy as a fulcrum to leap toward becoming a "Home AI Ecosystem Builder." Recently, SDMC was selected by Hang Seng Indexes Company Limited for inclusion as a constituent of the Hang Seng Composite Index, effective September 7, 2026, marking a key step in its capital market development. According to SDMC's interim results announcement released on August 20, 2026, the company achieved revenue of RMB 2.095 billion in the first half of the year, up 36.3% year-on-year; period profit was RMB 181 million, up 75.9% year-on-year; and gross margin increased by 3.5 percentage points year-on-year to 21.1%. Among this, digital video equipment revenue reached RMB 1.734 billion, up 40.5% year-on-year, accounting for 82.7% of total revenue; overseas revenue was RMB 1.947 billion, representing 92.9% of the total.
As a leading global provider of AI Home total solutions, SDMC continued to increase R&D investment during the reporting period, focusing on key areas such as the home AI agent Cedar, edge computing, smart terminals, and underlying storage synergy. This pushes the deep integration of cloud and home-edge solutions, establishing a complete technical architecture and commercial implementation pathway.
Edge AI Commercial Inflection Point Has Arrived: Home Scenarios Become the Core Deployment Ground
Driven by both technological iteration and market demand, edge AI is accelerating into a phase of large-scale commercial adoption. This inflection point is shaped by the concurrent maturity of three driving forces: computational cost, privacy compliance, and real-time responsiveness. The increase in AI chip computational density and the decline in costs have made it economically viable for home terminals to integrate 1-5 TOPS of local processing power, supporting high-frequency tasks like voice recognition and image analysis to run locally. Due to the unique nature of home environments, purely cloud-based AI applications face multiple challenges. Edge AI, by processing home data locally and combining it with the flagship intelligent experience of cloud AI, achieves an optimal balance between experience and security. The rigid demand for millisecond-level responses in applications like security alarms and fall detection also makes edge-side real-time decision-making an indispensable infrastructure.
In terms of market potential, the edge AI sector shows significant high-growth characteristics. According to data from Fortune Business Insights, the global edge AI market was valued at $35.81 billion in 2025, projected to climb to $385.89 billion (equivalent to RMB 2,617.1 billion) by 2034, with a compound annual growth rate of 29.9% from 2026 to 2034. This far exceeds the growth rate of the cloud computing market, as capital and technological resources rapidly shift toward edge intelligence. The industry chain exhibits a three-tier architecture, and edge AI solutions targeting smart homes, which address the coordination challenges of "multiple devices, multiple brands, and multiple protocols," represent one of the more noteworthy application directions.
Currently, smart homes are transitioning from "single-point intelligence" to "scenario coordination." The data silo model of independent APP control has become a major bottleneck hindering experience upgrades. The next phase's core is building a home AI hub with a closed loop of perception, decision-making, and execution, which must satisfy four conditions: local operation, multi-protocol compatibility, cross-brand interoperability, and AI-native design. SDMC's "AI Home" strategy precisely targets this strategic gap, reshaping the value chain of the home intelligent ecosystem through the fusion of edge AI and intelligent connectivity.
Decoding the "AI Home 1+2+X" Architecture: Building an Ecosystem Moat via Cloud-Edge-Device Synergy
SDMC's strategic upgrade essentially builds on its global leadership in Android TV terminals to execute a value leap from "software and hardware product solutions" to "AI Home total solutions." The core support for this transformation lies in two scarce assets it has accumulated: First, a B2B2C channel network covering over 80 countries and regions, serving more than 300 operator and retail brand customers as of June 30, 2026, reaching hundreds of millions of households and laying the foundation for large-scale commercialization. Second, a deep partnership with the Google ecosystem, with early positioning within Google's TV and AI ecosystems, providing a technological first-mover advantage for its AI strategy.
SDMC's "AI Home 1+2+X" architecture represents a system-level solution with "cloud-edge-device" synergy. Its core value lies in forming an ecosystem moat through technological integration, rather than through single-product innovation. "1" refers to the home AI agent Cedar, the self-developed "home brain." As of June 2026, Cedar has been further equipped with the Gemini 3.5 large model, continuously enhancing capabilities in visual perception, proactive services, identity recognition, and personalized memory. Its cloud-edge combined deployment architecture offloads certain functions to devices like AI smart TV boxes, supporting network self-maintenance and local offline AI customer service. During the reporting period, Cedar completed R&D and is now in the customer testing and commercialization phase. Strategically, Cedar has evolved from an interaction entry point into the "decision-making hub" for the home scenario.
"2" represents the two foundational pillars: intelligent connectivity and edge AI. The intelligent connectivity layer is centered on the Wi-Fi 8 AI optical gateway, which integrates tri-band Wi-Fi 8 with 10G optical access and, crucially, incorporates an AI NPU. This enables intelligent traffic scheduling, Wi-Fi Sensing wireless perception, and local AI inference collaboration. On the software side, the AI optical gateway, combined with prplOS's remote management and application lifecycle management capabilities, supports operators in large-scale device management and continuous feature upgrades, transforming the gateway into a programmable edge computing node. It is particularly important to emphasize that, beyond gateway products, SDMC's AI smart TV box is another key carrier of the home edge computing architecture. SDMC pioneered turning AI smart TV boxes with processing power into operators' AI Home hubs, further integrating the agent with home network devices, showcasing significant phased results at MWC 2026. Looking ahead, the company will continue to deploy edge AI devices like home computing centers to complete the home edge computing system. In essence, SDMC's AI smart TV box has evolved from a traditional streaming playback terminal into a home edge computing node that integrates local processing power, runs the Cedar agent, and coordinates whole-home devices.
"X" represents the continuously expanding terminal ecosystem matrix, which has grown from streaming boxes to include smart projectors, cameras, home service robots, and smart glasses. These terminals collaborate deeply with the home hub devices through Cedar to achieve scenario-based coordination. When a camera or other sensor identifies an anomaly, the AI Home system can judge based on home context and coordinate compatible devices for alerts and responses, thus forming a complete scenario loop from perception, to decision-making, to execution.
From Hardware Box Leader to Home AI Hub Operator: The Open Ecosystem Connector and Value Re-rating Logic
More importantly, SDMC's product innovation path fundamentally reshapes value from "functional terminal" to "AI hub." The Wi-Fi 8 AI optical gateway has evolved into a core node for home edge computing, characterized by built-in processing power (integrated AI NPU with millisecond-level decision latency), enhanced perception (Wi-Fi Sensing for unobtrusive monitoring), and programmability (supporting dynamic application deployment via prplOS). This transforms the gateway from a network connection device into an operationally valuable platform capable of continuous value-added services. In parallel, the AI smart TV box, leveraging its broad home coverage and high-frequency user interaction, serves as the carrier for the Cedar agent. It functions both as a standalone AI customer service and scenario automation engine, and collaborates with the gateway for computational synergy, enabling dynamic task allocation and load balancing at the edge. Cedar is extending from a single-device capability toward a cross-device AI service layer, offering value-added services such as network self-maintenance, local offline AI customer service, and scenario automation. This drives the operations model from "manual-driven" to "AI-driven."
SDMC's strategic insight lies in not pursuing a closed ecosystem but instead becoming a "connector" for mainstream ecosystems through a "dual-entry" strategy. On one hand, by supporting the Matter protocol, Cedar can control Matter-certified devices across brands, bridging different ecosystems and smart hardware for interoperability and coordinated action. On the other hand, deep integration with Google's business and capability ecosystems allows SDMC to deliver the "last mile" for customers and provide users with unique scenario definitions. This "dual-entry" strategy secures SDMC's irreplaceable ecological niche within the home environment.
Conclusion: The Rarity of the Strategic Position
The high scarcity of SDMC's strategic positioning fundamentally stems from its simultaneous possession of three core elements that are extremely difficult to combine in the home edge AI track: First, the B2B2C channel barrier serving over 300 operator and retail brand customers, constructing a moat for large-scale monetization. Second, comprehensive cooperation with the Google ecosystem, securing a core ecological position within user scenarios. Third, the "1+2+X" system-level architecture capability, achieving a qualitative leap from stacking individual products to whole-home intelligent coordination. At the critical juncture where edge AI moves from concept to implementation, SDMC, through the Cedar agent and its edge computing foundation, is transforming the home scenario into an operable, value-appreciating digital asset. Understanding this reconstruction of the business logic—from "selling terminals" to "selling scenarios"—forms the underlying rationale for assessing its investment value and value re-rating potential post-listing.