A new wearable device that looks like a simple beanie could soon change how people interact with computers. Developed by Silicon Valley startup Sabi, the prototype uses brain-computer interface (BCI) technology to convert a user’s internal speech into text, effectively allowing them to “type” using their thoughts.
According to a report by WIRED, the device is designed to be one of the least intrusive brain-tech wearables yet, avoiding the bulky, futuristic look of many experimental headsets. Instead, it blends into everyday clothing, making it more practical for daily use.
A Wearable That Listens To Your Thoughts
The beanie works using electroencephalography (EEG), a non-invasive method that reads electrical signals from the brain through the scalp. Unlike implant-based systems such as those being developed by companies like Neuralink, Sabi’s approach does not require surgery.
The goal is to detect “internal speech” – the words people think but don’t say out loud – and convert it into text on a connected device. If successful, this could allow users to communicate or control devices without speaking or typing.
To improve accuracy, the beanie reportedly includes tens of thousands of miniature sensors, far more than traditional EEG devices. This high-density sensing is designed to capture more detailed neural signals, helping the system better interpret what the user is thinking.
Why This Approach Matters
Brain-computer interfaces are not new, but most existing systems fall into two categories: invasive implants or bulky external hardware. Both approaches have limited mainstream adoption due to medical risks, cost, or usability challenges.
Sabi’s beanie takes a different path by focusing on comfort and accessibility. The device is designed to work out of the box without requiring daily calibration, which has been a major limitation for many BCI systems.
Early targets suggest typing speeds of around 30 words per minute, with the potential to improve as users become more familiar with the system.
This could open up new use cases, from accessibility tools for people with disabilities to hands-free computing for everyday users.
The Challenges Of Reading The Mind
Despite its promise, the technology faces significant hurdles. Brain signals vary widely between individuals, and even the same thought can produce slightly different neural patterns each time.
To address this, Sabi is developing a large-scale AI model trained on thousands of hours of brain data collected from volunteers. The aim is to identify patterns that correspond to internal speech across different users.
However, experts caution that “mind-reading” remains a complex and often overstated concept. Current systems can decode limited patterns or commands, but translating continuous, natural thought into text is still an evolving challenge.
Privacy And Ethical Questions
One of the biggest concerns surrounding this technology is privacy. Neural data is deeply personal, potentially revealing thoughts, intentions, and cognitive patterns.
Sabi says it is addressing this by encrypting data and working with neurosecurity experts to ensure safe handling. Still, the broader conversation around brain data rights and ethical use is likely to grow as such devices move closer to commercialization.
What Comes Next
The company is aiming to release its first consumer version of the beanie – and a cap variant – by late 2026.
If successful, the device could represent a shift toward more subtle, everyday brain-computer interfaces that integrate seamlessly into daily life.
For now, the idea of typing with your thoughts may still feel futuristic. But with devices like this beanie, that future is starting to look a lot more wearable—and a lot less intrusive.