

A Swiss startup just tested a battery-less brain implant that beams neural data through your skin using infrared light. If the tech holds up in chronic trials, it could reshape the entire brain-computer interface field.
Imagine a device inside your skull that reads your thoughts, beams them through your skin using infrared light, and never needs to be recharged. No battery. No wires poking through your scalp. No second surgery five years later to swap out a dead power cell.
That's what Ability Neurotech just tested in living humans for the first time.
The Swiss startup implanted its battery-less brain-computer interface (BCI) in patients undergoing brain tumor surgery at TUM University Hospital in Munich. Up to five patients participated. Each one had 20 to 30 minutes of neural recording while their skulls were already open for tumor removal. It's a clever piggyback strategy: no extra surgery, no added risk, just a chance to prove the tech works in the real world.
And the data suggests it does.
Most brain implants face the same annoying trade-off. You want high-quality signals? You need wires through the skull, which invite infection. You want wireless? You need a battery inside the implant, which means it's bulky and will eventually die.
Ability's approach sidesteps both problems. The implant sits beneath the scalp and records electrical brain activity using ECoG electrodes (thin grids that rest on the brain's surface, rather than needles that pierce into it). Then it sends that data through intact skin via a transcutaneous infrared optical link, basically a tiny laser communicating with an external receiver.
Think of it like a fiber optic cable, except the "cable" is your own skin.
The bandwidth is staggering: 50 megabits per second, with less than one sample lost per million. End-to-end latency clocks in at roughly 2 milliseconds. For context, that's fast enough to decode speech and movement in real time. The system can detect signals as faint as 0.18 microvolts while handling swings up to ±6 millivolts.
Because there's no battery, the implant can theoretically last as long as its materials hold up. No replacement surgeries. No recharging schedules. The external unit handles all the heavy lifting: power delivery, data reception, and AI-based decoding.

Leqembi just became the first Alzheimer's drug you can take entirely at home, from first dose to last. With nine-month clinic waitlists and infusion-center bottlenecks choking patient access, this FDA approval could reshape who actually gets treated.


Join thousands of biotech professionals who start their day with our free, daily briefing.
This is where the design gets especially interesting. In most BCI systems, the implant itself runs at least some of the software that translates brain signals into commands. That's a problem, because updating software inside someone's skull typically means another trip to the operating room.
Ability flipped the script. All decoding happens in the external unit. The implant is essentially a high-fidelity microphone for your brain; it captures everything and ships it out raw. The algorithms that interpret those signals live outside the body, where they can be upgraded, swapped, or improved without touching the patient.
It's like buying a TV that will automatically support every future streaming service, because the smarts are in the cloud, not the hardware. This "future-proof" architecture means the system could adapt to new tasks (speech, movement, cognitive monitoring) as AI decoding improves over time.
The choice of surgical setting wasn't random. Intraoperative neural recording during tumor removal is already standard practice. Surgeons routinely map brain activity to avoid cutting into regions that control speech or movement. By testing during these procedures, Ability got access to exposed brain tissue in a controlled clinical environment without subjecting anyone to surgery they wouldn't otherwise need.
The first patients were recorded under general anesthesia, validating that the hardware, optical link, and processing chain all work together in a live operating room. Next steps will involve conscious patients performing speech and motor tasks while the system records. That data will train the AI models needed for real-world communication.
The Munich study is just the opening act. Ability has already secured ethics approval in the Netherlands for a chronic implantation trial in ALS patients at UMC Utrecht, targeting Q3 2026. That trial, part of a consortium called INTRECOM (involving UMC Utrecht, TU Graz, and device partner CorTec), will be the first time Ability's fully implantable, wireless, battery-less BCI lives inside a person long-term.
The goal: restore communication and speech for people to use at home. Not in a lab. Not during a controlled experiment. In someone's actual daily life.
For patients with ALS or locked-in syndrome, who retain full cognitive function but have lost the ability to move or speak, this kind of technology isn't a cool gadget. It's the difference between being trapped inside your own body and being able to tell your family you love them.
Ability isn't operating in a vacuum. Neuralink has its high-density cortical implant and three FDA Breakthrough Device designations. Synchron threads electrodes through blood vessels with lower surgical risk and has already linked its BCI to Apple Vision Pro. Blackrock Neurotech supplies the Utah arrays powering much of academic BCI research.
All three are further along commercially. Ability doesn't yet appear in major industry rankings or analyst shortlists. It's the new kid, spun out of Geneva's Wyss Center for Bio and Neuroengineering in 2025 after roughly eight years and $50 million of development.
But its technical thesis is distinct. Where Neuralink optimizes for channel density and Synchron for minimal invasiveness, Ability is betting on raw data throughput and external intelligence. If the optical link proves stable over months of daily use, analysts suggest it could force a rethink of RF-based wireless BCIs that compress data on-chip and embed fixed decoders.
One commentator put it bluntly: if ABILITY's approach works chronically, it "raises a viability question for everyone who built around the problem it solves."
We're watching a Swiss startup, founded by a team including CEO Rotem Kopel and technical co-founders Arnau Espinosa, Jonas Zimmermann, and David Ibáñez, attempt something no one has done before: build a brain implant that's fully implantable, battery-free, and powerful enough to stream raw neural data at broadband speeds through your skin using light.
The first human test is done. The chronic trial is approved and coming. If the technology holds up outside the operating room, in real homes, with real patients, Ability won't just be another BCI company. It will have built the infrastructure layer that every future brain-computer application runs on.
The brain's data has always been there. Ability just found a way to let it out.
J&J's OTTAVA just became the first table-integrated surgical robot to earn FDA authorization, creating an entirely new device category. After a decade of delays, billion-dollar lawsuits, and a $900 million write-down, the fight for Intuitive Surgical's robotic surgery throne just got real.