BRENNEN

Industries / Frontier science

Brain-computer interfaces

Brain-computer interfaces (BCIs) are devices that read signals from the brain, or send signals into it, so that a person can work a computer or machine, or regain a lost ability such as moving, speaking, feeling touch or seeing.

1973

J. J. Vidal published "Toward direct brain-computer communication" in the Annual Review of Biophysics and Bioengineering, setting out a design for turning brain signals into digital inputs for computers. A 2024 review in Nature Reviews Bioengineering says the paper coined the term "brain-computer interface".

api.crossref.orgpar.nsf.gov

47

Milestones achieved across 7 domains.

15

To watch for: announced, not yet confirmed outside.

15

Open deliverables. Nobody has shipped them.

In one minute

Where it started

In 1968 Giles Brindley and Walpole Lewin made a blind patient see spots of light by stimulating the visual cortex, and in 1973 J. J. Vidal named the idea of a brain-computer interface. Recordings from the scalp let volunteers spell and move cursors from the late 1980s. In 1998 a man with locked-in syndrome became, by the count of a 2024 review, the first person with long-term implants to move a computer cursor with brain signals.

Where it stands

About 80 people had received implants for communication or movement by mid-2025, most of them in the US, by the count of a 2025 review; Neuralink, Paradromics, Synchron and Chinese teams such as Neuracle have implanted more people since. In research studies people with paralysis or ALS have written by imagined handwriting at 90 characters a minute and spoken through decoders at 62 to 78 words a minute. China approved its first implanted BCI for sale, Neuracle's NEO, in March 2026; in the US no implanted BCI for computer control is approved, though deep brain stimulators that adjust themselves were approved in 2025.

What is open

Much of what is known sits in single-lab papers, company posts and separate national registries. The open jobs below build shared tables (trials, harms, electrode lifetimes, rules in four countries), shared tests (phantom heads for headsets, gel blocks for insertion robots, simulators for decoders, vision encoders and stimulation controllers) and audits of consumer brain-data apps. None of them needs a device on a person or any patient's data.

Domains

AchievedWatch forOpen

Open in brain-computer interfaces

3 of 15 open deliverables

Open · Implants

Open soak-test data for films

A bench rig that soaks test pieces of the thin plastic films used to insulate flexible brain electrodes (polyimide, parylene C and liquid crystal polymer) in warm salt water, some with dilute hydrogen peroxide added to mimic the body's chemical attack on implants. It logs each piece's electrical leakage every hour until the piece fails, and publishes every reading.

Built with what exists

Comb-shaped test patterns of interleaved metal tracks, ordered from a commercial flex-circuit maker in polyimide, with parylene C added by a coating service; water baths held at 37 °C, 57 °C and 87 °C; phosphate-buffered saline and dilute hydrogen peroxide; a Raspberry Pi with a multiplexer and an analogue-to-digital converter to log leakage current; Python with SciPy to fit Weibull curves, a standard statistical model of time to failure. An AI coding agent can write and test the logging and fitting code.

Credit · unclaimed

Open · Wearable devices

One phantom, many headsets

A public scorecard of consumer and open-hardware EEG devices (headbands, headsets, EEG headphones and earbuds), each tested on the same phantom head: a gel or 3D-printed model that conducts electricity like the scalp, fed known signals from a signal generator. Each device is scored on its noise level, how well it recovers a known 10 Hz signal of set size, timing jitter against a light-triggered marker, and dropped data.

Built with what exists

A conductive gel or agar head phantom, or the 3D-printed conductive phantom described by the Heriot-Watt team; a two-channel signal generator or a microcontroller with a digital-to-analog converter; a photodiode or small solar cell for timing markers, as in the Maynooth method; Lab Streaming Layer to record every device on one clock; MNE-Python for analysis; an OpenBCI Cyton board as a low-cost reference. The advance is cheap 24-bit EEG amplifier chips such as Texas Instruments' ADS1299, used on OpenBCI boards, which a 2016 side-by-side test by Jérémy Frey found could stand in for a medical-grade g.tec amplifier in BCI tasks.

Credit · unclaimed

Open · Decoding signals

Decoder energy on real chips

An open table that runs the same five or more movement decoders (a Kalman filter, a linear filter, a small recurrent network, a spiking neural network and a small transformer) on at least three low-power chips sold today, and measures accuracy, delay and the energy each decode uses with a power meter. All runs use recordings already published from two monkeys, Indy and Loco, in the NeuroBench "primate reaching" task, so there is no new animal work and no human data. An implant has a small power budget, so these numbers decide which decoders could ever run inside one.

Built with what exists

NeuroBench's open code and primate reaching sessions; PyTorch; Google's open-source Coral Compiler v1.0 and the Synaptics Coralboard SL, a board with Google's Coral NPU (neural processing unit) that Google says is on sale worldwide from DigiKey and targets 10 mW average power; the Arduino Nano 33 BLE (Arm Cortex-M4) that NeuroBench itself used as its microcontroller baseline; a bench power monitor. Small, cheap AI chips with open compilers are the advance that makes this a desk job. The site's Note "An open compiler" covers the Coral Compiler.

Credit · unclaimed