The Exoskeleton Race: China's Robotic Legs Are Changing How We Move

Imagine strapping a robot to your legs and suddenly feeling like you could run forever. At CES 2026, this wasn't science fiction—it was the hottest trend on the show floor. Personal exoskeletons were everywhere, promising to transform how we walk, run, and climb.

From Factory Floor to Main Street

Exoskeletons have made the leap from industrial and medical applications to consumer products faster than anyone expected. The market is exploding from $500 million in 2025 to a projected $2 billion by 2030. Companies like WiRobotics, Sumbu, Ascentiz, and Dephy showcased ambitious designs, while established players like Dnsys and Hypershell refined their offerings.

These aren't the bulky mechanical suits from movies. Modern exoskeletons wrap around your waist like a belt, with hinged mechanical supports extending down to your thighs. They use AI to analyze your gait in real-time, providing robotic assistance for walking, running, squatting, and climbing stairs.

Only two companies currently sell consumer-ready exoskeletons: Hypershell and Dnsys, both Chinese startups founded in 2021. We put their flagship models—the $1,999Hypershell X Ultra and the $1,899Dnsys X1 Carbon Pro—through rigorous testing at London's Lea Valley Athletics Center.

The Numbers Don't Lie

Both companies make bold claims about performance benefits. Hypershell promises 42% lower heart rate, 20% less exertion when walking, and 63% increase in hip flexor endurance. Dnsys suggests their device can decrease power demand by up to 50%.

Our testing revealed these aren't just marketing numbers. During 400-meter jogs, the exoskeletons significantly reduced physical effort. One tester saw his heart rate drop from 174 bpm to 151 bpm with the Hypershell. However, the benefits varied dramatically between individuals—a fit, active 48-year-old found them more hindrance than help, while a 76-year-old with a titanium hip could climb hills without his usual rest breaks.

The technology clearly works, but the user experience differs dramatically between the two models.

A Tale of Two Exoskeletons

The Hypershell X Ultra impressed with its polished execution. Nearly silent operation, smooth assistance, and intuitive controls made it feel like a finished product. The carbon fiber construction and titanium components justify the premium price, delivering up to 1,000 watts of peak power with 18.6 miles of range per battery.

The Dnsys X1 Carbon Pro, while cheaper and lighter, felt unfinished. Every step produced "RoboCop sound effects" that quickly became tiresome. The exposed cables connecting the hip-mounted battery pack looked messy, and the assistance felt jerky—"like Woody from Toy Story," as one tester put it.

Both use AI to recognize terrain and adjust power output automatically, but Hypershell's implementation feels more refined. Dnsys boasts about AI developed by former DJI, Segway, and Xiaomi engineers, yet the execution doesn't match the pedigree.

The Sprint Test Surprise

In 60-meter sprints, both exoskeletons actually improved performance. The naturally faster runner clocked 11.25 seconds unaided but 10.63 seconds with the Hypershell. The slower runner saw even bigger gains, dropping from 12.03 to 10.51 seconds.

This wasn't expected. These devices are marketed for endurance and assistance, not speed enhancement. Yet the robotic boost translated into measurable performance gains, suggesting applications beyond their intended use cases.

Market Implications and Challenges

At roughly $2,000, these exoskeletons cost about the same as a high-end smartphone. The question isn't whether the technology works—it does—but whether consumers will adopt devices that make them look like cyborgs.

Chinese companies have a significant head start, but opportunities remain for Western manufacturers. The market is young, and user experience clearly matters more than raw specifications. Companies that can solve the noise, weight, and aesthetic challenges while maintaining performance could capture significant market share.

Regulatory questions also loom. How will these devices be classified? Will insurance cover them for medical applications? What happens when enhanced humans share sidewalks with unenhanced pedestrians?

The Bigger Picture

Personal exoskeletons represent more than assistive technology—they're human augmentation devices that blur the line between natural and artificial capability. As the technology improves and prices drop, we may see a future where robotic assistance becomes as common as smartphones.

The implications extend beyond individual users. What happens to public spaces designed for unassisted humans? How do we address the potential divide between enhanced and unenhanced populations? These questions become more pressing as the technology matures.