3 min readNew DelhiUpdated: Feb 7, 2026 08:54 AM IST A new type of Artificial Intelligence chip has been created by scientists that is capable of bending and flexing while at the same time performing a computing process. This could have a huge impact on future wearable technology. The work was highlighted in a recent report, which examined how flexible electronics are moving beyond sensors and into real-time AI computing.
According to the Nature journal, the research shows that it is now possible to build AI-capable circuits on flexible materials instead of traditional rigid silicon wafers. This would potentially allow wearable technologies such as smart watches, health patches, and even smart clothing to analyse the data directly on the body, rather than sending the data to a phone or remotely located computer.
Researchers were able to prove that neural networks, which form an integral part of the latest AI technologies, can function normally even if the chip is bent or curved. This marks a shift from earlier flexible electronics, which were mostly limited to simple signal detection or data collection tasks.
How the research works
The study explains that the flexible AI chip uses thin-film transistors arranged in a way that allows computation even when the material is under mechanical stress. It does not use a high-performance silicon chip but instead uses simpler parts, so that power consumption is lower, as well as more durable.
Although these computing powers are not high enough to equalise with smartphone processing techniques, they are more than adequate for the functions needed in wearable devices, such as recognising motion patterns, recognising alteration of vital signs, and filtering of data before storing and transmission. Researchers note that this approach would “significantly” reduce energy use in wearables. Since the device processes information locally, it does not need to transmit large amounts of raw data, which is often one of the biggest drains on battery life.
Why it matters for wearables
Today’s wearable devices still depend heavily on rigid chips, which limit how thin, soft or flexible they can become. As a result, some wearables are still bulky or uncomfortable for extended wear. However, there can be a solution in the form of a “bendable AI chip” which may be capable of moulding itself in accordance with the body.
Flexible AI systems may prove valuable in healthcare applications, too. Devices that are strapped to the body can analyse data such as heart rhythms, muscle movements, and so on, and can alert the individual without needing to be connected to the cloud. This may be important for healthcare-related applications where data privacy is an imperative.
Expand © IE Online Media Services Pvt Ltd