Source: Unsplash
Imagine your display refresh rate is measured not in hertz, but in neurons. Imagine your color palette is white dots on a black background, comparable to early Atari graphics. Imagine the undo button does not exist. Welcome to the interface challenge of this decade: brain-computer interfaces are becoming real, and the UX problem is a design emergency that almost nobody in the industry is taking seriously.
Brain-computer interfaces (BCIs) are no longer a science fiction concept. In 2026, Neuralink is beginning first human trials of Blindsight, its device that writes visual information directly into the brain's visual cortex to restore sight to the blind. Synchron has now implanted its Stentrode device in over 50 patients with paralysis across trials in the US and Australia, with zero serious adverse events. Paradromics completed its first-in-human recording of the Connexus device in May 2025, a chip smaller than a dime with 420 individual electrodes embedded directly into brain tissue. The global BCI market sits at approximately $3.75 billion in 2026 and is projected to reach $13.86 billion by 2035. Total BCI investment crossed $1 billion in 2024-2025 alone, with Synchron's $200M Series D round (total: $345M) representing a coming-of-age moment for the space.
The technology is arriving. What's not arriving nearly fast enough is the design thinking around it. And as someone who has spent eight years building product interfaces for some of the most complex systems in enterprise and healthcare, I want to tell you: the UX of brain-computer interfaces is the hardest design problem our generation will face.
What the Interface Actually Is
Most conversations about BCIs focus on the neuroscience and the engineering. The electrode density. The signal fidelity. The surgical precision. What gets almost no attention is the fundamental interface design question: how does a person interact with a system that communicates directly with their brain?
Let me make this concrete with Neuralink's Blindsight. The device works by capturing video through a camera, processing it, and then firing electrical signals into the visual cortex to create artificial perception. Early results show patients experiencing "phosphorous flashes" and basic geometric outlines of their environment. Resolution is described as comparable to early video game graphics. Basic. Pixelated. Monochrome at first.
Now here's where my designer brain starts asking questions that Neuralink's engineers aren't publicly answering. How does the user know the signal quality is degrading? What does a connection dropout feel like, and how is it communicated? When there's an error in the image processing pipeline, what does the user experience, and how do they know something is wrong versus just seeing their environment correctly? When there's a software update, what happens to their visual field during the transition?
These are not engineering questions. They are interaction design questions. And they don't have clean answers yet.
"UI/UX designers need to craft user experiences for something as intimate as a brain interface, especially as BCIs are combined with AI assistants and AR/VR goggles for immersive control. Brain product designers face neurological, biological, and ethical problems that all have to be solved simultaneously and from day one."
— Built In, "How to Design Products for the Brain"
This is exactly right. And exactly what the industry is not doing. The companies building BCIs right now are, almost uniformly, engineering-led organizations. The founders are neuroscientists and hardware engineers. The early hires are from deep-tech and medtech backgrounds. The design team, where one exists, is typically a small UX crew working on the companion app and the surgical dashboard. Nobody is asking the deeper question: what is the full interaction model when the human nervous system is part of the system?
The Three Hardest UX Problems in BCI Design
I want to break down three specific design problems that I think are going to define whether BCIs actually become products that people can live with, or remain sophisticated medical devices that patients tolerate.
- The Error State Problem: In every product I've designed, error states are some of the most important screens. What happens when something goes wrong? Does the user know? Can they recover? In a web app, a broken state means a page doesn't load. In a BCI, a broken state could mean visual distortions, phantom signals, or an absence of perception where there should be one. Designing the error language for these states, how to communicate them to the user without adding cognitive load or panic, is an entirely new design discipline that doesn't exist yet.
- The Consent and Control Problem: Every interaction design system has a user in control of their own experience. They can close an app, turn off a setting, walk away. With an implanted BCI, the interface is inside your skull. The control model becomes radically more complex. Who has authority to push updates? What happens if the company providing the firmware goes bankrupt? (Neuralink's own user agreement reportedly contains a clause acknowledging this scenario.) These are not paranoid edge cases. These are the fundamental permission architecture questions that no mainstream design framework has ever had to answer.
- The Mental Model Problem: When someone uses a new interface, they build a mental model of how it works. With BCIs, that mental model has to account for something unprecedented: a device that both reads from and writes to the human brain. Users need to understand, in plain terms, what the device perceives, what it acts on, and what it ignores. Without a clear mental model, users cannot make informed decisions about their own behavior. This is not just a UX problem. It's a fundamental issue of human agency.
The Precedent That Should Scare Everyone
I've written before on Medium about the mindset shifts product teams need to make when building AI-native experiences. One of the most important: when you design the interface, you design the relationship between the user and the system. Get that relationship wrong, and no amount of technical sophistication saves you.
BCIs are about to test that principle at a level no previous technology has approached. We have a precedent for what happens when powerful technology is deployed without adequate interaction design: social media. Platforms optimized for engagement without regard for psychological impact. Notification systems designed to pull attention rather than respect it. Algorithmic feeds that shaped behavior at scale without users fully understanding the mechanism.
Now imagine those same design failures, but the interface is not a phone screen. It's a device inside your brain that can modulate your sensory perception. The stakes of getting the interaction model wrong are not measured in lost productivity or screen addiction. They are measured in something much harder to quantify.
This is not a call to slow down BCI development. The medical case for these devices is genuinely profound. Restoring sight to someone who has been blind for decades. Allowing a paralyzed person to communicate at the speed of thought. These are not incremental improvements in quality of life. They are transformational. The work Synchron, Neuralink, and Paradromics are doing deserves real support.
But that support needs to include the design discipline to match the engineering ambition. And right now, that discipline is absent.
What Good BCI Design Actually Looks Like
I want to end with something practical, because practitioner thinking is the whole point of this blog. Here's what I think good BCI interaction design needs to prioritize, based on what I know about designing complex, high-stakes interfaces.
Graceful degradation, not silent failure. Every state of the system, including low-signal quality, partial failures, and interruptions, needs a designed response. The user always needs to know what is happening. This might mean audio tones, tactile feedback through a companion wearable, or a visual indicator in a paired AR display. Whatever the mechanism, silence is the wrong answer.
Progressive disclosure of complexity. New BCI users should not be handed a full configuration dashboard on day one. They need onboarding that builds their mental model gradually, in context, as they actually use the device. The first week of using a Blindsight implant should have a designed experience: what to expect, what different signals mean, how to communicate problems to the care team.
User-controlled update cycles. Software updates to a BCI device should never be automatic. The user, or their designated caregiver, should have meaningful control over when updates are applied, what they change, and what the expected experience will be afterward. This is basic user agency applied to an unusually high-stakes context.
At reloadux, we've been writing about how AI systems need to earn trust rather than assume it. The same logic applies here, amplified by orders of magnitude. A BCI that the user cannot understand, cannot predict, and cannot control is not an interface. It's an imposition.
The engineers building these devices are solving for the 2020s. The designers who will make those devices actually work for real humans are solving for the next fifty years. Right now, there are not enough of them in the room. That needs to change, and soon.
What's your take on this? Are BCIs a medical device story, a design story, or something entirely new? I'd love to hear from anyone working in neurotech, assistive technology, or human-computer interaction. Drop a comment below.
Sources:
1. Neuralink Blindsight: Visual Prosthesis Trials — neuralink.com/trials/visual-prosthesis
2. Synchron $200M Series D Funding — businesswire.com/news/home/20251106150841/en/Synchron-Raises-200-Million-Series-D
3. Paradromics First Human Implant — paradromics.com/news/paradromics-completes-first-in-human-recording
4. Brain-Computer Interface Market 2026 — precedenceresearch.com/brain-computer-interface-market
5. How to Design Products for the Brain — builtin.com/articles/design-products-brain-computer-interface
6. Brain-Computer Interfaces 2026: Neuralink, Synchron, and Real Progress — 3zebras.com/science/brain-computer-interface-2026-neuralink-synchron-update
7. Brain-Computer Implants: 3 Trends to Watch in 2026 — statnews.com/2025/12/26/brain-computer-interface-technology-trends-2026
8. Neuralink Blindsight Won't Deliver Natural Sight — spectrum.ieee.org/neuralink-blindsight