The State of Brain-Computer Interfaces in 2026
A grounded snapshot of the brain-computer interface field as it stands, past the headlines and the hype, for readers who want the real picture.
Brain-computer interfaces in 2026 are real, working, and narrow. Implanted systems let a small number of people with paralysis control cursors, type, and move robotic limbs by intention alone. Several companies are racing along different points of the invasive-to-noninvasive trade-off. None can transmit language between minds, and none are remotely ready for healthy, elective users.
Brain-computer interfaces attract two kinds of coverage: breathless and dismissive. Neither is accurate. The reality in 2026 is more specific and, in some ways, more impressive than either. This is a grounded snapshot, written as a companion to my plain-English primer on what a brain-computer interface is.
What is actually working
The clearest results come from implanted systems used by people with paralysis. In research settings, participants can:
- move a computer cursor and click using intention alone,
- type by imagining handwriting movements, faster than letter-by-letter selection,
- control a robotic arm well enough for everyday actions like drinking.
These are not demos. For the people involved, they are restored capabilities. That is the genuine state of the art, and it is extraordinary.
The central trade-off, and who sits where
Every approach lives somewhere on one axis: how close the sensor gets to the neurons. Closer means a cleaner signal and finer control, at the cost of more invasive surgery and more long-term risk.
| Approach | How it reaches the brain | Signal | Trade-off |
|---|---|---|---|
| High-channel cortical implant | Electrodes in the cortex (open surgery) | Highest resolution | Most surgical risk |
| Vascular implant | Threaded in via a blood vessel | Good, lower than cortical | Avoids open-brain surgery |
| Surface array | On the brain surface | High | Surgery, less tissue penetration |
| Non-invasive EEG | Worn on the scalp | Lowest resolution | Safe, no surgery |
Companies and labs are deliberately spread across this axis. Neuralink is pushing the high-channel-count cortical end. Synchron is pursuing a stent-based implant delivered through the blood vessels, trading some signal quality for a far less invasive procedure. Academic groups and others work the surface-array and non-invasive ends. There is no single winner, because they are optimising for different points on the same curve.
What is not working yet
Being precise about the limits is the whole value of a snapshot:
- No mind-to-mind language. Nothing transmits a sentence from one brain to another. Decoding specific trained intentions is a world away from reading free inner speech.
- No elective use. Healthy people have no reason to accept brain surgery, so the entire field is, correctly, medical for now.
- Stability and scale are unsolved. Implanted electrodes have to keep working for years inside living tissue, and decoding has to generalise. Both are hard, open problems.
Why a book about language cares
Speech is a workaround for not being able to share a thought directly. Brain-computer interfaces are the first technologies aimed at the layer beneath the workaround. Even in their narrow medical form, they make a question concrete that used to be purely philosophical: what is communication once the body is no longer the only channel?
That question is where the evolution of language points, and it is the heart of how AI is changing human language. I follow it all the way out, carefully, in Building Your First Brain, free for the first 1,000 readers.
Further reading
- Synchron and Neuralink publish updates on their respective trials; read them alongside peer-reviewed work.
- Stanford’s Human-Centered AI institute tracks neural-interface and AI research for a general audience.
What Is a Brain-Computer Interface? A Plain Guide
A brain-computer interface reads signals from the nervous system and turns them into commands. Here is how it works, what it can do today, and what it cannot.
How AI Is Changing Human Language
Large language models and brain-computer interfaces are quietly rewriting the oldest human technology. Here is what is actually changing, and what is not.
Will We Still Need Words?
If thought could move between minds directly, would language survive? An essay on what words are for, and what would be lost and gained without them.