Some US cities make hailing a taxi a gamble—you might get a car with no driver. Duh, robotaxis aren’t new. But what if Physical AI could take over your own car? That’s what CES 2026 showed us, and not only.
What is physical AI?
To get what Physical AI is, let’s take a fun, a bit nerdy dive. First, let’s rewind to the origin of the word “robot.” It comes from the Czech word “robota,” meaning “forced labor.” When you hear the word “robot,” you probably think of the future, but the first industrial robot, Unimate, showed up in the 1960s. Like other robots at the time, Unimate was rule-based—it followed strict instructions to perform repetitive tasks with precision and speed, but it lacked flexibility.
Fast forward to today, and we’re seeing the rise of Physical AI. It’s all about training robots to learn from real-world or simulated experiences using AI and machine learning. In short, Physical AI lets robots and devices make decisions and act in the real world, not just in virtual spaces. Below, I’ll walk you throw how Physical AI can change your driving experience.
Ford AI assistant
At CES 2026, the Gadget Flow team came across the Ford AI assistant, which the American automaker calls an “intelligent thread”, a name that makes sense once you see how it works in real life.
Picture a trip to a lumber yard with stacks of firewood in front of you, from full cords to half cords to loose bundles, while your truck waits for a load. You want an answer, not guesswork or rough math. You pull out your phone, snap a photo of the stack, and ask how much will fit in your truck bed. The assistant reads the image, figures out the volume, and turns that pile into clear answers like bundle count or a portion of a cord. All based on your Ford truck bed size and tie-down setup.
I like the app-first approach from Ford because it makes the innovation accessible to more drivers. You don’t need a new truck to see where Ford AI goes next. The app brings smart tools and personal help to the Ford you already own. By 2027, Ford plans to integrate the AI assistant into Ford and Lincoln vehicles.
Mercedes-Benz MB. DRIVE ASSIST PRO
NVIDIA is teaming up with Mercedes-Benz to bring the future of driving to life. They’re providing the AI, software, and the powerful computing platform for Mercedes’ next-gen driver assistance systems. At CES 2026, NVIDIA’s CEO, Jensen Huang, shared a bold vision, saying, “we imagine that someday a billion cars on a road will all be autonomous.”
With MB. DRIVE ASSIST PRO, Mercedes-Benz is combining driving assistance with navigation to make driving safer and smoother. A simple button press lets the car navigate through city streets with SAE Level 2 support, from parking to the final destination. Plus, with their cooperative steering, the car can make steering changes at any point without turning off the system.
The other day, I was talking to a chauffeur who drives for an ambassador from a South Asian country. He was telling me how he does the same route every day—back and forth—and how it’s just so repetitive and boring. I couldn’t help but think that something like the MB. DRIVE ASSIST PRO could make a difference for people like him. It might not eliminate the routine, but it could at least take some of the mental load off during long drives.
MB. DRIVE ASSIST PRO will hit the US market later in 2026, though it’s already available in China.
Parting thoughts
Physical AI isn’t some distant concept anymore—it’s coming for our cars, our everyday routines, and the way we interact with machines. From Ford’s smart assistant helping you pack a truck to Mercedes’ MB. DRIVE ASSIST PRO easing long drives, the tech feels real and practical. I can’t wait to see how it changes the road ahead, and I’d like to try these innovations myself once they hit the market.