Scientists from the research group from Tsinghua University and Peking University in China have successfully developed an artificial intelligence (AI) chip capable of bending and stretching, marking an important step forward in the field of smart wearable device technology.
This innovation is expected to change the way health and exercise tracking devices will work in the future, while significantly reducing energy consumption.
According to a report published in the British journal Nature, researchers have proven that AI integrated circuits can be manufactured on flexible materials, instead of traditional hard silicon.
Thanks to that, devices such as smartwatches, health patches or smart clothes can process data right on the user's body, instead of having to send data to phones or external servers.
The breakthrough point of the study lies in the fact that the artificial neural network still operates stably even when the chip is bent.
Previously, flexible electronic devices mainly only undertook the role of sensors or simple data collection.
The integration of real-time AI computing capabilities opens up a completely new direction for flexible electronics.
Flexible AI chips are built based on thin membrane transistors, arranged in a structure that allows maintaining computing activity when the material is loaded or bent.
Instead of using high-performance silicon chips, the research team chose simpler components, which help reduce power consumption and increase mechanical durability.
Although the processing power of the chip cannot be compared with the processor in smartphones, it still well meets the essential needs of wearable devices.
Specifically, the chip can recognize motion, detect changes in biological indicators and filter data before storing or transmitting.
Another major advantage is local data processing. When the device does not need to send large amounts of raw data out, battery consumption is significantly reduced, while increasing the ability to protect personal information.
Currently, most wearable devices still depend on hard chips, making it difficult for them to achieve the thinness, softness and flexibility needed for long-term wear.
A curved AI chip is considered the solution to this limitation, helping the device automatically adjust its shape according to its body, bringing a more comfortable feeling to users.
In the field of healthcare, flexible AI also opens up many new potentials. Devices attached to the body can monitor heart rate, muscle activity or other biological indicators in real time and issue warnings immediately, even when not connected to the internet.
This is especially important for medical applications, where accuracy and data security are always put first.
