Tether Evo, the Artificial Intelligence (AI) and neuroscience research initiative of Tether , has published three new peer-reviewed research papers on Brain-Computer Interfaces (BCIs). The core purpose of this advancement is to create a universal AI brain-computer interface that carefully decodes speech, vision, and music across different individuals. Two of the papers were designed by Tether Evo in collaboration with the University of Rome Tor Vergata (UniTOV).
One model, many brains. ?
— Tether (@tether) August 4, 2026
Tether Evo just published 3 new peer-reviewed papers showing how a single BCI model can generalize across different people across 3 key areas:
Speech: Translating neural data to text in minutes
Vision: Reconstructing images from primate spiking data…
Basically, research based on three papers indicates that a unified model can learn to assess brain activity across people without starting from each individual. Tether Evo is actively solving the problem of a huge number of people in terms of providing them with a collective solution of speech, vision, and music. Tether has shared this news through its official social media X account.
Tether Evo Pushes the Future of AI-Powered Neural Interfaces
The developmental step of Tether Evo is going to solve one of the biggest hurdles in brain-computer interface (BCI) research, given that every brain produces slightly different signals. Tether Evo is Tether’s frontier technology division, committed to the intersection of biology and machine intelligence. Tether Evo builds first-local, high-performance systems designed to strengthen individuals and secure personal autonomy in a growing centralized world.
The three peer-reviewed studies cover three different angles of the brain under a single model. For that purpose, cross-subject decoding of human neural data for speech brain-computer interfaces has taken at the Journal of Neural Engineering. This model provides support to those people who have lost their senses, such as the sense of vision, hearing, and speaking, due to any injury or accident. This shows how a single model can be trained across various people at once.
Unveiling AI Model That Reconstructs Images and Decodes Music
Investigators from Tether and UniTOV unified to record brain signals from macaques viewing thousands of images and rebuilt what the animals were seeing directly from activity. The accuracy of this model is 70% with 200 milliseconds of neural data. This model generates a plausible reconstruction capturing its shape, color, and content.
In the same vein, through music, researchers recorded brain scans (fMRI) from five people as they listened to 540 songs spanning 10 genres. Tether believes that nothing is more private than your own information. Tether ensures people’s privacy by keeping their information secure and safe with the utility of an on-device AI stack. This action required no permission on any device.
