A BCI company at an XR expo — and how we met XPANCEO
We took BROADMIND to Singapore for the 9th AWE Asia in August 2023. It was the right room to be in — for a reason that has nothing to do with headsets.
August 2023BROADMIND
BROADMIND in the official AWE Asia 2023 Sponsors & Exhibitors lineup (courtesy AWE Asia).
When people hear that a brain-computer interface company spent two days at an augmented reality conference, the question is usually some version of: what were you doing there?
The honest answer is that XR has an input problem, and every input solution so far has been a compromise. Controllers are precise but clumsy. Hand tracking is natural but exhausting over a working session. Eye tracking tells the system where you are looking, not what you intend to do. Each of these measures a movement. None of them measures an intention directly.
That is the gap EEG fills, and it is why we were in Singapore.
The room
AWE Asia 2023 ran 30–31 August at the Max Atria, Singapore Expo, the ninth edition of the regional chapter of Augmented World Expo, organised locally by Sunrise International. The numbers the organisers published afterwards give a sense of scale: delegates from 31 countries and regions, 80+ speakers, 60+ exhibitors, thousands on the floor and more than 160,000 viewers on the livestream.
Forbes has called AWE "XR's Most Essential Conference," and the community around it goes back to 2010. What that means in practice is a room where the people building optical stacks, display modules and spatial operating systems are standing next to the people deciding which of those get deployed in a factory or a hospital. It is a commercially serious crowd, not a demo day.
What was actually on stage
Two things from the programme stayed with me.
The first was Dr. Mark Billinghurst (University of South Australia) on empathic tele-existence — systems that let a user physiologically experience being in another person's body. It is a research direction that only makes sense if you accept that the boundary between sensing a body and sensing a mind is thinner than the interface industry usually assumes.
The second was Alvin Wang Graylin (HTC) on the metaverse's risks, framed against what he calls digital globalisation — the idea that the same technology that could unify could also fragment, and that AI is the accelerant in both directions. Dr. Yu Yuan, then president of the IEEE Standards Association, spoke to the same theme from the standards side: AI and the metaverse are not two roadmaps, they are one.
Around them: SK Telecom and Deutsche Telekom on the global rollout of the ifland platform, Tipatat Chennavasin on what it actually takes for an XR startup to survive, and Dr. Yujia Gao (National University Health System, Singapore) on Holomedicine — mixed reality as a clinical tool. The Auggie Awards closed the show, with Immersal taking Best in Show.
Read that list again and notice what is missing: almost nobody on that stage was talking about input. The industry has spent a decade solving display, optics, tracking and comfort, and comparatively little time on the channel through which a person actually decides to do something.
Why we were in the exhibitor hall
BROADMIND appeared in the official AWE Asia 2023 Sponsors & Exhibitors list, alongside Niantic, Lenovo, XPANCEO, Dispelix, Immersal, Sunrise and around sixty others.
What we put in front of people was a different proposition from the rest of the floor: not a display, not a tracking system, but a non-invasive EEG headset with unrestricted access to raw data and an open-source-friendly software path. The pitch to an XR developer is not "control your computer with your mind" — that phrase sells a demo and disappoints a developer. The pitch is a latency budget, a calibration cost, and a documented data format.
Three questions came up repeatedly, and they are the right three:
How long does calibration take, and can I skip it? Every XR developer has been burned by a setup step that ruins the first five minutes of a session.
What happens when the user moves? Headsets move. Electrodes on a moving head produce artefacts, and artefacts are where naive pipelines fail.
Can I get the raw signal, or only your features? If a team cannot see the signal, they cannot debug it, and they will not build on it.
Those questions shaped our hardware and SDK priorities far more than any roadmap document did.
The uncomfortable number
There is a figure the BCI research community has lived with for years: roughly up to 30% of users show poor or no reliable control with standard BCI paradigms. It is not a hardware defect — it is a mismatch between a fixed decoding approach and brains that vary. In an XR context that number is fatal. A spatial interface that fails for three users in ten is not a product.
This is why AWE Asia was useful beyond the leads it generated. It clarified the standard we have to meet. XR sets the bar at works on the first try, for everyone in the room, without a lab technician. That is a much harder bar than a research demo, and it is the correct one.
The people in the next booth
One of the more useful conversations of the two days was with XPANCEO, the Dubai deep-tech company building smart contact lenses — display, biosensing and zoom inside a lens. They were a silver sponsor of the same edition, so we spent most of the show as neighbours.
On paper the two products could not be more different. A lens sits on the eye and puts pixels in front of it; a headband sits on the skull and listens to what is behind it. But the argument underneath is identical, and neither of us had to explain it twice: the bottleneck in spatial computing is no longer the display, it is the interface. Their answer is to make the display disappear into the eye. Ours is to remove the hand from the loop altogether.
That is a rare thing to find at a trade show — a company working a completely different part of the same problem. XPANCEO has since raised at a valuation north of a billion dollars for its lens work; we have kept in touch, and the conversation from Singapore is still the one we return to.
What we took home
Three things, none of them a partnership announcement:
Input is the open problem in spatial computing, and the industry knows it. The conversation is no longer whether neural input belongs in the stack, but where in the stack it sits.
The consumer and the developer want different products. Consumers want an experience that works out of the box. Developers want raw access and documentation. Building for both means one device, two surfaces — and not confusing them.
A booth at a generalist expo is a translation exercise. You get about fifteen seconds to explain why a headband is relevant to someone whose job is optical engineering. If you cannot do it in fifteen seconds, the problem is your explanation.
We went to Singapore to be seen by an industry that is not our own. We came back with a sharper definition of the bar we are building against — and a clearer sense that the most interesting interface of the next decade may not be the one you wear on your face.
BROADMIND builds non-invasive EEG hardware with unrestricted raw-data access and an open, developer-friendly software path. If you are building spatial interfaces and want to test neural input, get in touch.
BROADMIND in the official AWE Asia 2023 Sponsors & Exhibitors lineup (courtesy AWE Asia).