Will Meta's Brain2Qwerty Help You To Speak Your Mind Without Typing?

So, here's the tea of what took place in Spain. 35 volunteers sat down with rapt concentration typing simple sentences into a keyboard. Doesn't sound like a big deal. But, what was simultaneously happening was that machines were capturing every flicker of neural activity their brains were emitting when they were typing. Why is that? It's because of something Mark Zuckerberg's Meta is spearheading. And it's called Brain2Qwerty. Not the most elegant name, but it may catch on, who knows?

This is one of those BCIs (Brain-Computer Interfaces) that aim to decode sentences from neural activity that's recorded without someone actually typing words. That might be helpful for those who have speech and mobility issues. Plus, creatively, during interrogations, someone could hook someone up to these interfaces to find out what they're thinking and get confessions. Could neural data be able to reveal emotional states?

And that may be why those volunteers were typing sentences. That way, the system could ascertain the brain patterns linked to each keystroke. Based on that, some reverse-engineering could take place.

How's it going with Brain2Qwerty so far? It's said to be able to decode typed sentences at rates that some might think only belonged in a Nolan movie. While that may sound cool, it's said to be as big and expensive as an MRI machine. Plus, this BCI scanner is said to be able to only work in a shielded room where the Earth's magnetic field is dampened. If the room wasn't dampened, chances are the brain signals wouldn't be detected. And just like MRIs, head or body movements make the signals go kaput for the BCI scanner.

So, right now, this may not entirely be commercially available or ready yet.

Though, this could be interesting for India. Locomotor disabilities are said to account for close to 45% of all disabilities. And about 140 people per 100,000 in India were said to have speech disabilities. Maybe, formal physical therapy isn't as accessible and affordable as it should be. A scanner as advanced as the one Meta's working on may not be immediately feasible for widespread adoption in India. Could there be simpler hardware that could still improve baseline communication? 

Maybe, the AI could be finetuned to understand Indian languages. That would mean that some language-specific datasets may have to be created. That could really change the way speech therapy is conducted in India. Could something like this become multimodal? Imagine a Brain2Qwerty-esque scanner collabing with some eye-tracking software or that mouse for a mouth tech. If there was something like that, maybe that could help a person move around in a wheelchair. That'd be pretty interesting.

Perhaps, some of these BCI capabilities could even be used for VR or AR headsets. Type with your mind if your hands are busy. All sweet, but technical and financial landmines strewn ahead, beware.

Something, like this, is super deep even in a DeepTech ecosystem. Does India really even have a BCI realm yet? Probably, only lawyers are familiar with BCI, but for other reasons.

Would some VC, who might usually focus on verticals, like FinTech or SaaS or e-commerce, want to focus on this domain? For even DeepTech investors, this might be a new strain. For reference, maybe, someone coming out with this kind of tech might cite Elon Musk's Neuralink or the fact that Meta's working on a similar project. Of course, some might lose faith, keeping in mind how much funding Meta invested in the Metaverse, but, maybe, all hope is not lost for the Metaverse. Maybe, an impact investor might see some appeal with potential societal returns. Helping the disabled to communicate? That could work. Super high upfront costs, though.

Of course, the scanners that Meta is using are really heavy and require specialized and shielded rooms. Maybe, over time, those costs might drop, but it might be a really really long time before miniature versions of these devices are out. So, maybe, they may not be available D2C, but they could be sold to hospitals and similar healthcare facilities down the line.

How are you going to speak your mind?