DeepX CEO Proposed Physical AI Solutions

Kim Nokwon outlined a shift toward high-efficiency hardware to address data center infrastructure constraints.

Updated on Sept. 23, 2026 in Semiconductors

Bold flat-color editorial illustration of a large copper-colored semiconductor wafer, evoking the precision of high-efficiency hardware development. — DeepX CEO Kim Nokwon has unveiled a strategy for 'physical AI,' focusing on high-efficiency semiconductor hardware to bypass existing data center infrastructure bottlenecks. AI Illustration. Upload story photo >

Live Poll

Do you believe physical AI solutions will successfully address global data center power and infrastructure constraints?

DeepX CEO Kim Nokwon recently introduced a strategy for physical AI, aiming for efficiency levels 100 times higher than traditional data centers. This approach seeks to mitigate power, heat, and cost limitations currently facing the artificial intelligence industry.

Why it matters

Rising demand for AI has pushed data center infrastructure to its limits, resulting in critical shortages of power supplies, transformers, and printed circuit boards. Physical AI offers a localized alternative that reduces reliance on these strained resources.

Physical AI aims to achieve efficiency levels 100 times higher than those seen in current data center operations. This technology focuses on low-power semiconductor design to resolve persistent challenges regarding power consumption, thermal management, and production costs.

The players

Kim Nokwon

He serves as the CEO of DeepX and is a proponent of physical AI solutions for hardware constraints.

DeepX

The company specializes in developing low-power AI semiconductors designed to address heat and cost challenges.

Tekedra Mawakana

As the co-CEO of Waymo, she is scheduled to deliver a keynote address at the upcoming CES 2027 conference.

The details

DeepX is developing low-power semiconductors that integrate directly into consumer products, bypassing the massive infrastructure required for large-scale data centers. By moving processing to the edge, the company aims to bypass the current bottlenecks in power supply and hardware procurement.

September 21, 2026: Kim Nokwon attended a CTA media event in San Francisco.
September 23, 2026: DeepX announced its CEO attended the industry event.
2027: DeepX will exhibit new low-power semiconductors at CES 2027.

The Tech Race

This move toward physical AI reflects a broader industry pivot away from centralized, power-hungry processing toward decentralized, high-efficiency local hardware. It follows a pattern set by the industry-wide struggle to scale cloud-based AI infrastructure in the face of rising energy consumption.

Consumers may soon see AI-powered products that operate faster and more reliably without requiring constant connectivity to energy-intensive cloud servers. This transition could lead to longer battery life and reduced hardware costs for everyday devices utilizing advanced AI features.

The takeaway

Physical AI shifts the burden of processing from central hubs directly onto the hardware users carry in their daily lives. By prioritizing power efficiency, this transition aims to make AI features more sustainable and cost-effective as the technology becomes ubiquitous.