I am not bullish on brain-computer interface

https://marginalrevolution.com/marginalrevolution/2023/06/why-i-am-not-entirely-bullish-on-brain-computer-interface.html

by elsewhen • 3 years ago

5 9 3 years ago

marginalrevolution.com

dexwiz• 3 years ago

Not bullish, you mean bearish?

blamarvt• 3 years ago

“Not bullish” implies either neutral or bearish, I think.

aksss• 3 years ago

The main advantage of the direct brain-computer interface seems to be speed. But extreme speed is important in only a limited class of circumstances, many of them competitions and zero-sum endeavors, such as sports and games.

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.

gregjor2• 3 years ago

Calling bullshit on this now, purely based on Musk's track record. People continuing to bet on a horse that repeatedly stumbles in the back stretch doesn't make the horse a winner; it makes the bettors stupid.

brucethemoose22• 3 years ago

Neuralink is not the only horse in this race. Probably not even the favorite.

But its the most famous for sure.

gregjor2• 3 years ago

Yeah, and as far as I know the FDA is one regulatory agency Musk hasn't tried to circumvent, flip off, and get sued by yet, so green fields for him.

m463• 3 years ago

I liked ctrl-labs and thought they had great promise... until facebook came and acquired it (boo).

brucethemoose22• 3 years ago

BCIs (and maybe genetic intelligence enhancement?) are needed to counterbalance the rise of AI, IMO.

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.

NalNezumi• 3 years ago

It's too early to tell, but given current technology I'd say BCI is a hard egg to crack.

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

I am not bullish on brain-computer interface

Related Stories