Hassane El-Khoury does it again, though his role in the interplay among power, sensing, computing, and connectivity has reversed this time around. In 2019, as the newly appointed CEO of Cypress Semiconductor, he spearheaded the sale of his company to power and sensing player Infineon. As a result, the German chipmaker deepened its presence in compute and wireless connectivity designs.
Fast forward to 2026, El-Khoury takes onsemi, a supplier of power, sensing, and control semiconductors, to foray into the compute and connectivity worlds by snapping up Synaptics in an all-stock deal valued at approximately $7 billion. However, besides compute and connectivity, there is an AI touch all over this deal.
In a way, this deal is more about edge AI than anything else. “By adding Synaptics’ differentiated edge AI compute franchise and strong portfolio of human-machine interface and wireless connectivity solutions, onsemi is expected to extend its capabilities beyond power and sensing to intelligent systems,” onsemi said in its press release.
Though onsemi seems to place greater emphasis on physical AI, a nascent technology that mostly refers to embedded AI in devices like drones, robots, and autonomous vehicles. Physical AI marks the integration of AI models with hardware in sensors, motors, and actuators so they can perceive their environments, understand spatial relationships, and perform work in the real world.
View All This also unwraps the story behind onsemi’s largest acquisition to date.
What onsemi wants with Synaptics
The Scottsdale, Arizona-based onsemi has a market valuation of $46.17 billion. It’s a leading supplier of power semiconductors and the largest vendor of image sensors to the automotive market. However, besides its forte in automotive and industrial, onsemi is currently positioning itself to become a major player in AI data center power.
And, while onsemi is well-positioned in the energy grid for data centers, Synaptics’ technology will help extend its reach to intelligent edge, allowing the company to provide integrated, system-level solutions for physical AI applications in AR/VR smart devices, autonomous driving, and humanoids. For that, El-Khoury has outlined four technology pillars—sense, decide, act, and adapt in real time—to extend a larger AI opportunity beyond the AI data center and into edge applications.
He is confident that Synaptics’ edge AI computing, wireless connectivity, and human-machine interface (HMI) technologies, combined with onsemi’s power management and sensing expertise, will constitute these four pillars to capture the $30 billion physical AI opportunity. “The shift toward physical AI will require power, sense, connected compute, and control to work together seamlessly,” El-Khoury said.
Industry watchers broadly agree that this deal is about edge AI and physical AI. “It’s a clue about where AI is going next,” enterprise AI strategist Sanjeev Bode wrote in his LinkedIn post. “The next phase of AI will not only be about big AI models, big GPUs, and big data center bills… it will be about machines that can sense, understand, connect, and act in the physical world.”
The deal will also accelerate onsemi’s evolution from a power and sensing company to a supplier of systems for edge AI and physical markets. “The combination of onsemi’s strength in power, sensing, and control with Synaptics’ AI compute, wireless connectivity, and HMI technologies is going to create a market leader in what is to be known as the physical AI realm,” El-Khoury said.
Synaptics: The end of an HMI era
Synaptics was founded by semiconductor industry luminaries Federico Faggin and Carver Mead in 1986. Faggin, best known for designing the first commercial microprocessor, the Intel 4004, also founded Zilog in 1974. Mead devised a new methodology to divide the increasingly complicated chip design process into logic, circuit, and layout, and to separate them from the manufacturing process.
Synaptics’ efforts to replicate biological functions in silicon eventually led to the creation of HMI chips, widely used in laptop touchpads, touchscreen technologies, display drivers, human presence detectors, fingerprint biometrics scanners for smartphones and video, and far-field voice technology for cars and smart home devices.
Over the past couple of years, Synaptics has become a prominent player in edge AI by combining its strength in HMI with AI-native compute and connectivity. Take, for instance, its Astra platform, which combines purpose-built AI processors and NPUs for multimodal intelligence with wireless connectivity spanning Wi-Fi, Bluetooth, and GPS. It comes along with a full open-source software stack for rapid deployment.
It’s the end of the road for an HMI pioneer and a new start for a power-and-sensing technology specialist in the edge AI corridors. It’ll be interesting to watch how onsemi integrates Synaptics’ edge AI and HMI technologies into its silicon offerings. Another thing to see is how Synaptics’ largely consumer technologies mesh with onsemi’s automotive and industrial portfolios.
Synaptics, when contacted by EE Times, declined to comment on the news due to regulatory restrictions. The deal, subject to regulatory approval, is expected to close in mid-2027.
Read also:
[Synaptics Makes a Natural Progression to Edge AI](https://www.eetimes.com/synaptics-makes-a-natural-progression-to-edge-ai/)
[Synaptics and Google Pair Astra with Open-Source ML](https://www.eetimes.com/synaptics-and-google-pair-astra-with-open-source-ml-to-drive-ai-in-iot/)
[Google Open-Sources NPU IP, Synaptics Implements It](https://www.eetimes.com/google-open-sources-npu-ip-synaptics-implements-it/)