The global push toward artificial intelligence has hit a physical wall. While the initial wave of AI development focused heavily on massive data centers, the industry now faces a critical bottleneck: the limitations of cloud-based processing. Current AI deployments rely on centralized servers in data centers that struggle with high power-consumption, significant latency, and insufficient computing density. For applications requiring real-time interaction—such as autonomous robotics, smart wearables, medical devices, or smart home appliances—sending data to a remote cloud and waiting for a response is no longer viable.
The world is also becoming increasingly untethered. Devices must operate without physical data or power connections, relying instead on batteries and wireless signals. This creates a congested wireless environment where multiple devices compete for the same unlicensed spectrum, often leading to a degraded consumer experience. In this landscape, the challenge of shifting responsibility between parties has emerged as a major hurdle for manufacturers. When a device fails, customers often find themselves caught between different silicon vendors—one for the processor and another for connectivity—none of whom take full responsibility for the integrated solution.
To solve these challenges, the industry is shifting toward Agentic AI, where intelligence moves to the deep sensor edge. This evolution requires a new class of silicon that is human-aware, contextually aware, power consumption conscious and thermally efficient, all while delivering high-performance inference natively on the device. Human-interface analog and sensory technologies must also be integrated efficiently at the far edge of the network. Synaptics is leading this transition, reinventing its internal structure and product roadmap to deliver the unified foundation required for this next era of physical AI.
Edge AI Strategy: Intelligence at the Deep Sensor Edge
We have anchored our growth strategy in the guiding principle that AI inference must occur at the far end of the edge—where data is actually sourced, learned from and ultimately intelligently utilized. This deep sensor edge prioritization allows for latency-optimized and power-efficient experiences that cloud models cannot replicate. By embedding AI capabilities directly into the hardware, we are enabling devices to respond to their environment in real-time, making them more intuitive and efficient.
At the heart of this strategy is the Synaptics Astra™ platform, a market-leading family of AI-native microprocessors and microcontrollers. While many industry incumbents have attempted to secure market share by building walled garden software environments or proprietary toolkits, we have taken a more customer- and developer-friendly and focused approach by embracing an open-source architecture. This developer-friendly strategy acknowledges that AI models are evolving too rapidly for any single company to maintain a closed ecosystem. By providing open-source compilers and engaging with leading partners like Google and their research divisions, we allow the global developer community to innovate on its hardware.
A significant technical achievement within this strategy is the Synaptics Torq™ neural processing architecture, developed in collaboration with Google Research. This architecture enables Synaptics to run complex models, such as Google’s Gemma 3, natively on its processors. This capability ensures that we stay ahead of the curve as AI models become more sophisticated and decentralized.
The strategic shift to the Astra platform unlocks a range of new opportunities for us to invest in innovation and expand our capabilities. By delivering advanced AI inference directly at the edge, we are positioned to foster deeper collaboration with developers and partners, accelerate product development, and drive growth across multiple market segments. This approach enables Synaptics to continually enhance our technology stack, broaden our reach into emerging applications, and support our customers with tailored solutions that adapt to evolving industry needs. As we build on these strengths, we are poised to play a pivotal role in shaping the next era of edge AI, fueling our ongoing growth and leadership in the field.
Business Alignment: Primed to Win via Unified Solutions
Compute and connectivity (Wi-Fi, BT/BLE, Thread, UWB, GNSS/GPS) are inseparable in an untethered world. Yet many manufacturers face challenges when trying to integrate disparate components. These manufacturers are forced to source a processor from one vendor and a wireless chip from another, leading to high engineering costs and support challenges. We solve this challenge by delivering comprehensive system solutions that integrate high-performance processing with best-in-class wireless connectivity. Our competitive advantage is clear: while peers may offer processors, they often lack the sophisticated wireless IP that Synaptics brings to the table.
A prime example of this unified strategy in action recently occurred with a major Japanese home appliance manufacturer. While a competing silicon firm presented the customer with static PowerPoint slides of a theoretical solution, the Synaptics team demonstrated a working prototype running on the Synaptics Astra platform. In a matter of days, we integrated a Japanese voice-inference model into our system, allowing the appliance to understand and execute voice commands locally with low latency. This agility demonstrated that we don’t just sell chips; we provide a springboard for customers in the form of solutions to reach the market faster.
At Synaptics, we also know that monolithic System-on-Chip (SoC) devices must soon integrate an MCU with Wi-Fi 7, Bluetooth 6.0, and Thread. The versatility of these connectivity protocols is essential for the next generation of IoT devices, which must function reliably within increasingly congested wireless environments. Integrating these protocols onto a single die provides an efficient, low-power, and cost-effective bill of materials (BOM) for manufacturers.
To ensure maximum market penetration, we employ a dual strategy. While we push the integrated Astra ecosystem, we also offer host-independent connectivity SoCs that can be paired with non-Synaptics processors. Our platform-agnostic strategy enables Synaptics to engage in the wider Edge IoT market, even when customers choose alternative compute platforms.
Furthermore, our SKU map strategy leverages internal IP blocks to rapidly assemble chips that range from high-performance tiers to cost-optimal implementations. This allows our customers to scale their software investments across an entire product portfolio, further increasing the stickiness of the Synaptics ecosystem.
Physical AI and Robotics: Designing the Robotic Intelligence Core
At Synaptics, we view the robotics market as a vast spectrum, ranging from the intelligent vacuum cleaners already found in homes to the advanced humanoids of the near future. Our goal is to provide the robotic intelligence core, the underlying technology that allows these machines to sense, move, and react with human-like precision.
A breakthrough area for Synaptics is tactile sensing and dexterity. In collaboration with partners like Grinn, we have demonstrated robotic hands that use dozens of touch controllers to mimic the human sense of touch. These sensors enable a robot to distinguish between different materials and modulate its grip force—for example, knowing to hold a ceramic cup with more pressure than a delicate paper one. This level of precision is made possible by ML/AI algorithms integrated directly into the touch controllers at the deep sensor edge.
The complexity of modern robots requires a massive amount of data to be transported from sensors to the main processor with minimal latency. Synaptics solves this through our interface bridge products, which serve as high-bandwidth data highways within the robot’s body. By leveraging our core IP from the video interface and VR markets, we ensure that a robot’s brain processes sensory input in real time, which is essential for safety and performance.
We have already secured a significant foothold in this space, actively sampling silicon for pilot humanoid builds with a leading North American market leader. These pilots, scheduled for delivery by the end of 2026, incorporate double-digit touch controllers and high-bandwidth data interfaces from Synaptics.
There are distinct differences in how these markets ramp. Consumer applications, such as smart TVs and household robots, typically see faster adoption cycles. For example, we recently secured a win with a Tier 1 consumer OEM for vision-based gesture control in smart televisions. In contrast, industrial robotics require higher precision, mission-critical reliability, and fail-proof performance, leading to longer pilot-to-production cycles that the company projects will reach scale in 2027 or 2028. However, once these industrial solutions are deployed, they tend to have much longer lifecycles—often five to ten years—providing long-term stability for the business.
The Path Toward an AI-Native Future
After nine months of transformation under my leadership as the new CEO, Synaptics has emerged as a focused powerhouse in the Edge AI and IoT sectors. We are a solution provider defining the future of how humans interact with machines.
Our momentum is evident in our financial performance, with Core IoT product revenue increasing 53% year-over-year in the recently reported fiscal quarter, driven by the strength of our wireless connectivity portfolio. As we prepare for major industry events like Embedded World, we are set to unveil a roadmap that pushes the boundaries of Agentic AI, robotics, and augmented reality.
With upcoming announcements regarding our partnership with Google and the launch of new robotic developer kits, we are establishing Synaptics as the essential architect of the Edge AI era. By providing purpose-built silicon that integrates compute, connectivity, and sensing, Synaptics is ensuring that the next generation of intelligent devices will be more powerful, more efficient, and more human-aware than ever before. The transition from cloud-dependent AI to localized, physical AI is underway, and Synaptics is leading the charge.