Electronic implants designed to give blind patients back some sight are entering a new phase after two decades of mixed results. Earlier retinal devices have been abandoned, and researchers are now testing wireless systems that send visual information straight to the brain, alongside a subretinal prosthesis that has already restored central vision in a small study.
Tim Jackson, a professor of retinal research at King’s College London, describes the field’s trajectory and its open questions, including whether future devices could add thermal imaging, optical zoom or night vision.
- Key Point: The pioneering Retina Implant AG chip and the Argus II system were both dropped in 2019 after failing to deliver solid outcomes.
- Key Point: A Belgian company, ReVision Implant, plans to begin its first human clinical trial in October 2026 using a chip placed in the brain’s visual processing area.
- Key Point: The Prima prosthesis restored functional central vision in 81% of patients in a small study and is cleared for use in 30 European countries.
- Key Point: Stem cell therapies, still in early human trials, could eventually outperform electronic implants if they regenerate damaged retinal cells.
Why the first generation fell short
Most early work targeted the retina, the light-sensitive layer at the back of the eye whose failure can wipe out sight entirely. The German subretinal chip from Retina Implant AG demanded roughly eight hours of surgery, including drilling an opening in the skull behind the ear to hold a power receiver wired to the implant beneath the skin. What patients saw amounted to flashes of light only loosely matching the shape of what they faced, and many chips degraded over time, possibly because of rapid eye movement or corrosion from saline fluid inside the device.
The Argus II paired a camera with a chip driven by 60 electrodes. It won clearance from the U.S. Food and Drug Administration in 2013 but likewise failed to produce convincing results, and both systems were discontinued in 2019.
A wireless route into the brain
Research that survived those setbacks has moved in a different direction: beaming power and visual data wirelessly from an external camera to a chip implanted where the brain handles sight. Unlike retinal implants, this approach could also help people whose blindness stems from damage to the optic nerve, the pathway linking the retina to the brain, rather than from the retina itself.
Jackson cautions that it remains unclear how much detail such systems can deliver, since the natural processing carried out by the retina is lost. ReVision Implant, based in Belgium, is pursuing this method and intends to launch its first human trial in October 2026.
Prima’s results and the ethical trade-off
A second device, Prima, uses special glasses to transmit images and power by infrared signal to a chip surgically placed under the retina. It was tested in patients with dry age-related macular degeneration, one of the leading causes of blindness worldwide. In a small study, the prosthesis brought back functional central vision in 81% of cases, with participants able to read letters, numbers and words. Prima is authorized for use in 30 European countries.
Its commercial appeal is broader than that of earlier devices because macular degeneration affects far more people than the inherited retinal diseases the first implants targeted. That reach also raises an ethical question. Patients with inherited conditions have often already lost most of their sight, whereas those with macular degeneration typically retain good peripheral vision, giving them more to lose if surgery involving genuine risk goes wrong.
Because all these systems carry visual information electronically, extra functions can be layered on. Prima already offers optical zoom, and devices able to capture wavelengths beyond the visible range could one day provide thermal imaging and night vision. A prototype with that capability has been tested in primates, though Jackson notes such features will matter only once basic artificial vision matures far beyond its current level.
The stiffest competition may come from biology rather than electronics. Stem cell therapies aimed at regenerating retinal cells lost to disease are in the earliest stages of human testing and, like the first implants, have accumulated disappointments and face serious obstacles in getting the cells to survive and integrate into the eye. If they succeed in producing working replacement cells, they could over time surpass electronic implants entirely as a way to restore sight.
For now, the near-term test is whether brain-directed implants can clear human trials and whether Prima’s early success holds up at larger scale. Both paths will be judged on the same measure the first generation failed: whether patients gain usable vision, not just flashes of light.