CES: How Physical AI is Transforming Global Industries

The integration of AI into physical systems is developing as companies move beyond chatbots and digital assistants to address real-world challenges in manufacturing, healthcare and transport.

This transition is coming as advances in computing power and robotics enable AI systems to interact with and manipulate physical environments.

The semiconductor industry has responded to this evolution by developing specialised processors that can handle the computational demands of physical AI applications.

These developments arrive as manufacturers face persistent labour shortages and healthcare providers seek solutions to serve aging populations and underserved communities.

In this context, the Consumer Electronics Show in Las Vegas has highlighted the transition in AI from digital applications to physical world implementations, with major technology firms unveiling systems designed to bridge the gap between software capabilities and practical deployment.

Nvidia launches computing platform for AI development 

Nvidia demonstrated its expansion beyond graphics processing with Project DIGITS, a computing system the size of a laptop capable of running LLMs with 200 billion parameters.

“AI will be mainstream in every application for every industry. With Project DIGITS, the Grace Blackwell Superchip comes to millions of developers,” says Jensen Huang, founder and CEO of Nvidia in a blogpost.

“Placing an AI supercomputer on the desks of every data scientist, AI researcher and student empowers them to engage and shape the age of AI.”
The system aims to provide access to AI development tools for individual users and small organisations.

The company also unveiled Cosmos foundation models, a set of software tools that generate synthetic training data for autonomous robots and vehicles and this approach reduces the cost of collecting real-world data needed to develop autonomous systems.

The company additionally introduced AI Blueprints and these process data from various sources including images, videos and documents to generate business insights.

Semiconductor firms compete for AI market 

AMD presented its Ryzen AI Max processors for gaming computers and AI 300 series chips for business systems and Intel demonstrated its Core Ultra 200V processors for enterprise computing.

Meanwhile, Ambarella, which produces chips for edge computing - processing data near its source rather than in centralised facilities - introduced processors capable of handling twelve simultaneous video streams while running multiple AI models.

John Deere expands autonomous vehicle range 

John Deere also revealed its second-generation autonomy technology for industrial vehicles and this system incorporates cameras and computer vision - technology that enables machines to interpret visual information, to operate vehicles without human intervention.

The company has also expanded the technology from tractors to include orchard vehicles, quarry trucks and electric-powered lawn mowers for commercial use and this development addresses labour shortages in agriculture and construction sectors.

Furthermore, Caterpillar demonstrated autonomous electric machinery for mining and construction operations, focusing on site efficiency and safety improvements.

Healthcare firms target access challenges 

OnMed also presented its CareStation system, a self-contained medical facility that provides remote consultations through high-definition cameras and diagnostic equipment.

This system aims to serve communities without local medical facilities, which affect one in ten Americans.

Meanwhile, the healthcare sector demonstrated technologies for aging populations, such as adults over 65 who represent 16% of the US population but account for 37% of healthcare usage.

Market projections from AARP, the US advocacy organisation for older adults, indicate spending on age-related technology could reach US$96tn in three decades.
Additionally, Withings presented Omnia, a mirror system that measures vital signs and generates health reports through visual scanning technology.

Autonomous vehicle development accelerates 

The transport sector exhibited progress in autonomous driving systems - such as Kodiak, which reported 50,000 miles of self-driving truck operations in August 2024.

Continental also announced plans to manufacture Aurora Driver, an autonomous driving system, by 2027.

Zeekr unveiled its RT robotaxi developed with Waymo, the autonomous driving unit of Google parent company Alphabet and the vehicle will undergo testing in the US this year.

Also, Hesai presented sensors that provide thirty times higher resolution than current systems and Lidar, which uses laser light to measure distances, enables vehicles to detect obstacles and navigate autonomously.

The company projects unit costs below US$200, potentially enabling wider adoption in mass-market vehicles.

Robotics firms target industrial applications 

Agility Robotics demonstrated its Digit humanoid robot designed for warehouse operations and this machine performs physical tasks including moving containers.

The company plans to construct a manufacturing facility capable of producing 10,000 units annually.
Schaeffler further showcased applications for humanoid robots in logistics operations.

The demonstrations included robots performing repetitive warehouse tasks and moving heavy items, addressing labour requirements in industrial settings.

Finally, Realbotix presented Melody, an open-source humanoid robot with vision systems and micro-cameras that enable eye contact, speech and movement tracking capabilities.

Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.

Discover all our upcoming events and secure your tickets today.

Technology Magazine is a BizClik brand