Wearables are getting more capable every year, but most still depend heavily on a nearby smartphone to handle serious processing. That reliance may soon fade. A new breakthrough from researchers at Tsinghua University and Peking University suggests that future wearables could think for themselves, without sending data to a phone or cloud servers.
The research team has created an ultra-thin, flexible AI chip that is slimmer than a single strand of human hair. Even more impressive, the chip can be folded and bent thousands of times without breaking. Known as FLEXI, the technology could open the door to smart patches, health monitors, and other wearables that operate independently.
Instead of acting as simple sensors that pass data elsewhere, wearables powered by FLEXI could run artificial intelligence directly on the device. That means faster responses, better privacy, and far lower power consumption. According to the researchers, this approach could dramatically change how wearable technology is designed and used.
Inside a chip that bends without breaking
What truly sets FLEXI apart is not just its intelligence, but its physical structure. Based on findings published in Nature, the chip is built on a thin plastic film using low-temperature polycrystalline silicon circuits. This flexible foundation allows the entire system to twist, stretch, and bend while keeping the AI circuitry fully functional.
Because everything sits on a pliable surface, FLEXI can conform closely to the human body. This makes it ideal for wearable applications like smart medical patches, skin-mounted sensors, or lightweight health monitors. Unlike rigid silicon chips, FLEXI does not crack or lose performance when placed under stress.
To prove its durability, the researchers subjected the chip to extreme mechanical testing. FLEXI survived more than 40,000 bending cycles and continued working even when folded to a radius as small as one millimeter. Throughout these tests, the chip maintained stable performance, showing that flexibility does not come at the cost of reliability.
Strong performance with minimal energy use
Flexibility alone would not matter if performance suffered, but FLEXI delivers impressive real-world results. During health monitoring trials, the chip detected irregular heart rhythms with an accuracy of 99.2 percent. It also tracked everyday physical activities like walking and cycling with 97.4 percent accuracy.
Equally important is efficiency. FLEXI uses less than one percent of the power required by conventional chips performing similar tasks. This ultra-low energy consumption is critical for wearables, where battery size is limited and frequent charging is inconvenient.
Cost is another major advantage. According to reporting from TechXplore, the chip is expected to cost under one dollar per unit when mass-produced. This affordability could make advanced AI wearables accessible far beyond premium devices, potentially transforming consumer health tech.
What this means for future wearables
With FLEXI, wearables may no longer need to rely on smartphones for processing or connectivity. Health monitors could analyze data instantly on the body, improving speed and reducing privacy risks associated with cloud computing. Audio wearables could process voice commands locally, while lightweight augmented reality glasses might handle visuals and gestures without external hardware.
The research team plans to expand the chip’s capabilities by integrating additional sensors and increasing system complexity. These next steps could push flexible AI wearables closer to everyday use, especially in medical and fitness applications where comfort and accuracy are critical.
Beyond wearables, FLEXI fits into a broader wave of innovation in mobile hardware. Engineers are already exploring technologies such as chips that use controlled surface vibrations to make phones thinner and faster, as well as new cooling methods using synthetic diamond materials to improve performance in compact devices.
As these advancements come together, flexible AI chips like FLEXI could play a central role in shaping the next generation of smart technology. Devices that once depended on phones or cloud servers may soon operate fully on their own, quietly and efficiently, right on the human body.
