Imagine you’re wearing a hat that can understand your thoughts. At first glance this idea seems absurd and not close to reality.
To one’s surprise, Silicon Valley startup Sabi is developing a brain wearable device that can decode a person’s inner speech into words on a computer screen.
This brain-reading beanie is based on a technology called brain-computer interface, which provides a direct communication path between the brain and an external device.
Using a dense array of 70,000 to 100,000 miniature EEG sensors woven directly into clothing, Sabi is developing a non-invasive thinking cap that will translate imaginary words into text.
CEO Rahul Chhabra said, “Given that high-density sensing, it pinpoints exactly what neural activity is happening and where. We use that information to get more reliable data about what a person is thinking.”
Initial winter-hat designs are expected to be released in late 2026. According to Chhabra, the company also plans to design a more streamlined baseball cap.
Khosla, founder of Khosla Ventures, one of Sabi’s investors, said, “The biggest and worst application of BCI is whether you can make your computer talk by thinking. If you want a billion people to use BCI to access their computers every day, it can’t be invasive.”
Initially the company is working on a starting typing speed of 30 or more words per minute. This speed is often slower than most people would like, but this speed will eventually improve over time depending on the time users spend with the cap.
Here’s the catch! There are some drawbacks to wearable devices designed for the brain. For example, it is possible that the sensors cannot pick up strong nerve signals coming from the brain because of the layers of skin and bone. In contrast, surgically implanted devices have the advantage of receiving stronger signals.
In addition, BCI models also rely on AI models to translate neural signals into real-time commands. To power the interface, Sabi has trained a “Brain Foundation Model” using 100,000 hours of neural data collected from 100 volunteers.
