Princeton University to Lead AI-Driven Semiconductor Development for Wireless Technologies
Princeton University announced on August 6, 2025, that it will spearhead a collaborative government-industry initiative focused on using artificial intelligence to develop advanced semiconductors. These chips are essential for next-generation wireless networks, satellite communication, autonomous vehicles, and smart healthcare devices.
Kaushik Sengupta, professor of electrical and computer engineering at Princeton, will direct the project. His team aims to automate the design of microchips used in radio-frequency (RF) wireless communication—the technology that enables devices to exchange information wirelessly and interact with the physical environment.
Funding and Strategic Importance
The National Semiconductor Technology Center, a public-private consortium managed by the National Center for the Advancement of Semiconductor Technology (Natcast), provided nearly $10 million in grant funding. Natcast CEO Deirdre Hanford emphasized that adopting AI in RF design is vital for maintaining U.S. leadership in technology innovation. She highlighted that AI accelerates research and keeps the country at the forefront of communication infrastructure development.
Challenges in Wireless Chip Design
Wireless devices—from laptops to car sensors and satellites—depend on chips capable of transmitting high-speed, low-latency, and energy-efficient signals. Designing these specialized wireless chips is costly and demands highly specialized skills. Sengupta explains, “They are fundamentally handcrafted.”
Unlike chips for computers and data centers, which benefit from extensive automated design processes, wireless chips face unique challenges. They must operate under overlapping forces and unpredictable environments. Each phase requires input from experts in various fields, leading to long development cycles, high costs, and limited room for innovative design solutions.
AI-Driven Design: A New Approach
Traditional chip design starts with functional circuit knowledge and gradually adapts to meet modern requirements. Sengupta’s team flips this approach by beginning with performance demands and working backward to discover optimal circuit designs using AI. This method often results in chip architectures that defy conventional intuition but deliver superior performance.
Graduate students Emir Ali Karahan and Zheng Liu, guided by Sengupta, presented their AI-assisted chip designs at the 2022 IEEE International Microwave Symposium, winning the event’s top award and gaining significant recognition in the wireless technology community. Their related paper also earned the 2023 Best Paper Award from the IEEE Journal of Solid State Circuits.
Collaboration and Advanced AI Techniques
Mengdi Wang, another professor of electrical and computer engineering at Princeton, will contribute expertise in artificial intelligence and machine learning. The team plans to leverage two main AI techniques:
- Reinforcement Learning: Known for producing AI that excels at strategic games like Go, this technique will help automate complex design decisions.
- RF Diffusion Models: Inspired by methods used in Nobel Prize-winning protein design chemistry, these models will assist in generating innovative chip architectures.
The collaboration includes experts from the University of Southern California, Drexel University, Northeastern University, and industry partners such as RTX, Keysight, and Cadence. Senior leaders from Qualcomm, Skyworks, Texas Instruments, Nokia Bell Labs, Ericsson, and Maury Microwave will provide strategic guidance through an advisory board.
This initiative is set to accelerate semiconductor innovation for wireless communication by integrating AI directly into the design process. For IT and development professionals interested in AI applications across industries, exploring AI-driven automation and design can offer valuable insights into emerging technologies shaping hardware development.
For more on AI courses and training that align with these advancements, visit Complete AI Training.
Your membership also unlocks:
