I am not bullish on brain-computer interface
marginalrevolution.com
marginalrevolution.com
This statement seems very short-sighted, like the infamous 640k comment.
When talking about extreme speed, we’re talking about reducing latency. If I could type this comment with my brain instead of the slow interface that is my thumbs + phone keyboard, that would be a huge win for a seemingly mundane activity, and highly valuable. Writing code would be much faster, for instance. There are plenty of valuable uses for direct brain to computer interface, even simple functions, that don’t involve competition and zero-sum endeavors. It’s like someone saying graphics cards are only useful to gamers.
Being someone that have been working on it for the last two years all I can say is that it's too noisy and unreliable. There have been some excitement with using ML/DL as improved decoders but it's still hampered by myriad of problems for it being a product, or even better than existing interface devices.
Working with those devices suddenly gives you a massive dose of scepticism to neuroscience papers that use hose devices for experimental verification of their theories.
Advancement in (hardware side) ECoG is the only way I can see it get better but that's still far fetched
But its the most famous for sure.
If we are going to be bombarded with data, we need brains and interfaces that can keep up with it. Otherwise how can we even make intelligent AI related decisions and development.