## Does Ability Neurotech's Infrared BCI Solve Wireless Bandwidth for Chronic ECoG?

Geneva-based Ability Neurotech has begun its first-in-human intraoperative study in Germany, recording neural signals from brain tumor surgery patients in 20-to-30-minute sessions using a 128-channel [ECoG](https://bciintel.com/glossary/ecog) implant that streams uncompressed data at up to 50 megabits per second through the skin via infrared laser. The company's approach — battery-free, percutaneous-connection-free, and transmitting raw neural data without downsampling — directly addresses three of the most persistent failure modes in chronic implantable [brain-computer interface](https://bciintel.com/glossary/brain-computer-interface) design. A separate year-long chronic trial at University Medical Center Utrecht in the Netherlands, recruiting people with [amyotrophic lateral sclerosis (ALS)](https://bciintel.com/glossary/als), is planned for late 2026. If that trial proceeds on schedule, Ability will be among a small cohort of companies — alongside [Neuralink Corp](https://bciintel.com/companies/neuralink) and [Synchron](https://bciintel.com/companies/synchron) — running fully implanted wireless BCI systems in longitudinal human studies simultaneously.

*Disclaimer: Results reported here are from early-stage feasibility and intraoperative studies, not controlled clinical trials. No efficacy conclusions can be drawn at this stage.*

---

## The Engineering Bets: Optical Link, 128 Channels, No Battery

Most wireless implantable electronics use radio frequency transmission. Ability chose a different path: an infrared laser optical link that streams neural data through the skin to an external headpiece. The system samples 128 channels at 30,000 times per second each and transmits the full, raw recording — no compression, no onboard spike sorting, no downsampling — at up to 50 Mb/s.

CEO Rotem Kopel told IEEE Spectrum the company evaluated multiple transmission approaches before committing to the optical link, driven by one constraint: get raw recordings out of the body intact, and push decoding to external hardware.

The tradeoffs are real. Moving that volume of data generates heat and demands power. Engineers had to seal electrode connections inside a hermetically sealed case while routing an infrared path through tissue of varying thickness, vascularity, and hair coverage — all while tolerating imperfect alignment between implant and headpiece. Kopel says the technology took approximately 10 years to develop, followed by roughly 18 months of bench, durability, and validation testing before the first human use.

The implant is powered wirelessly through induction — no internal battery that could exhaust or require explantation. This matters for chronic use: battery replacement surgeries carry cumulative procedural risk, and eliminating that intervention is a meaningful clinical advantage if the inductive link proves reliable over multi-year timescales.

---

## What the ECoG Architecture Means vs. Penetrating Electrodes

Ability's [electrocorticography](https://bciintel.com/glossary/electrocorticography) [electrode array](https://bciintel.com/glossary/electrode-array) sits on the brain's surface rather than penetrating cortex. This positions the device differently from intracortical approaches used by Neuralink (thin-film penetrating electrodes) and the BrainGate consortium's Utah arrays. Surface ECoG generally trades single-unit spike resolution for a larger spatial footprint with less mechanical tissue disruption — a relevant consideration for longevity studies.

Maitreyee Wairagkar, a project scientist in the Neuroprosthetics Lab at UC Davis, offered independent context to IEEE Spectrum: "Preserving the information available in raw neural data through transmission is also useful for decoding purposes, since precious data is not lost due to transmission limitations." She added that battery-free design combined with the optical link "sounds very promising for chronic ECoG recordings."

Wairagkar also contextualized the bandwidth figure: other fully implantable BCIs — both ECoG and penetrating-electrode designs — use high-bandwidth architectures, with some already exceeding 50 Mb/s. Ability is not alone at this spec, but its combination of uncompressed transmission, battery-free power, and surface ECoG form factor is uncommon.

---

## Trial Structure and Clinical Timeline

**Intraoperative study (Germany, active):** Brief 20-to-30-minute recording sessions in brain tumor surgery patients. Later phases will include up to five conscious patients performing speech and movement tasks.

**Chronic trial (Netherlands, planned late 2026):** University Medical Center Utrecht is recruiting ALS participants for a year-long study. Participants will receive the implant, train with it at home, and undergo repeated brain-signal recordings as part of a project to restore speech for paralyzed individuals.

The staged design is methodologically sound. Intraoperative safety data typically precedes chronic implantation, and Kopel says the company's near-term goals follow that logic: evaluate the implantation procedure and device safety first, then test computer control, then pursue real-time speech decoding.

The harder problem, as Wairagkar put it, comes after implantation: "It is important that the underlying signal quality obtained from the device is maintained over long periods." Software can be updated remotely, but signal drift from electrode-tissue interface changes or neural plasticity is a hardware problem with no remote fix. ECoG's surface placement may offer some advantage here — without penetrating cortex, the foreign body response is reduced — but multi-year chronic ECoG data in implanted humans remains sparse.

---

## What This Means for the BCI Industry

Ability Neurotech's entry into human trials opens a new axis of differentiation in the implantable BCI market: transmission architecture. For the past several years, competitive positioning has clustered around electrode count, decoding accuracy in bits per second, and regulatory pathway speed. The optical link raises a distinct question — whether radio frequency transmission is actually the right default for high-channel-count chronic BCIs — that now has a human trial attached to it.

For the ALS community specifically, the Utrecht recruitment is significant. ALS progression is relentless, and year-long trials in this population carry urgent practical weight: participants may have limited communication windows. The pressure on Ability to demonstrate decoding utility — not just device safety — within the trial period is correspondingly high.

The device longevity question extends beyond Ability. Every implantable BCI company faces the same long-term support problem Kopel acknowledged: patients may remain dependent on an implant long after the company has changed, been acquired, or disappeared. The source text notes Kopel addressed this through a relationship the article does not fully detail — a disclosure gap worth watching as the company moves into chronic trials.

---

## Key Takeaways

- Ability Neurotech has begun its first human intraoperative study in Germany using a 128-channel ECoG implant with infrared optical data transmission at up to 50 Mb/s.
- The system transmits raw, uncompressed neural data — no onboard spike sorting or downsampling — sampled at 30,000 Hz per channel.
- A year-long chronic trial at University Medical Center Utrecht, targeting ALS patients for speech restoration, is planned for late 2026.
- The implant is battery-free, powered by inductive wireless charging, eliminating battery replacement surgery.
- Independent expert Maitreyee Wairagkar (UC Davis Neuroprosthetics Lab) assessed the architecture as promising for chronic ECoG use, while noting other implantable BCIs also achieve comparable or higher bandwidths.
- Long-term signal quality maintenance — not software updates — is the central unresolved challenge for multi-year BCI use.
- This trial is early-stage feasibility only. No efficacy or clinical outcome data are available.

---

## Frequently Asked Questions

**What is Ability Neurotech's BCI and how does it work?**
Ability Neurotech's implant is a surface ECoG device with 128 channels that records broadband neural signals sampled at 30,000 times per second per channel. It transmits raw data wirelessly through the skin via an infrared laser optical link at up to 50 Mb/s to an external headpiece, which relays data to an external processor for decoding. The implant is powered wirelessly by induction — there is no internal battery.

**How does the optical link differ from standard BCI wireless transmission?**
Most implantable wireless electronics use radio frequency (RF) transmission. Ability's infrared optical link operates by passing a laser signal through skin tissue, which the company chose specifically to accommodate the high data rate generated by transmitting 128 channels of uncompressed neural data. According to UC Davis researcher Maitreyee Wairagkar, some other implantable BCIs also achieve 50 Mb/s or higher via various wireless architectures — the optical approach is uncommon but not unique at this bandwidth.

**What clinical trials is Ability Neurotech running?**
As of August 2026, Ability has an active intraoperative study in Germany recording from brain tumor surgery patients in 20-to-30-minute sessions, with later phases targeting conscious patients performing speech and movement tasks. A separate year-long chronic trial at University Medical Center Utrecht in the Netherlands is planned for late 2026, enrolling ALS patients for speech BCI research. NCT numbers were not available in the source material.

**Why use ECoG instead of penetrating intracortical electrodes?**
ECoG arrays rest on the brain's surface without penetrating cortex, reducing mechanical tissue disruption and the associated inflammatory foreign body response. This may support better signal stability over multi-year implant periods, though long-term chronic ECoG data in implanted humans remains limited. The tradeoff is reduced access to single-unit spike data compared to intracortical designs.

**What are the main obstacles to long-term BCI use for ALS patients?**
According to Wairagkar, the critical issue is maintaining signal quality from the electrode-tissue interface over years — a hardware challenge that cannot be addressed by software updates alone. Additional concerns include decoder calibration as neural signals change over time, software maintenance, and company continuity risk: patients may remain dependent on an implant long after a startup has changed or been acquired.