Discussion
Interesting article
https://news.berkeley.edu/2024/10/02/researchers-s...
So they recorded the whole brain, modelled it into a simulation and the bloody thing acts like a fly
https://news.berkeley.edu/2024/10/02/researchers-s...
So they recorded the whole brain, modelled it into a simulation and the bloody thing acts like a fly
Gary C said:
Interesting article
https://news.berkeley.edu/2024/10/02/researchers-s...
So they recorded the whole brain, modelled it into a simulation and the bloody thing acts like a fly
https://news.berkeley.edu/2024/10/02/researchers-s...
So they recorded the whole brain, modelled it into a simulation and the bloody thing acts like a fly
Remember that episode of Black Mirror where they take a copy of your brain and turn it into a digital assistant that understands your every requirement? But inside it’s as conscious as you are, so they torture it into obedience.
This is step one on the path to that future.
I’ve watched and read too much sci-fi.
The fly world have built a 'connectome' of the D.melanogaster nervous system. It really is a tour de force of genetics coupled to functional biology. Now, as this has been extended to computer modelling, it can be used to predict all sorts of neurological things. But can it be done in mammals - maybe? In Humans, obviously not. There is the whole AI question though with humans. AI scores top marks in the 'Torrance test'. I might get 50/100.
Gary C said:
Thats the real question isn't it
could they in theory do the same to a human brain
Trusty old D. melanogaster was sequenced in 2000, H. sapiens followed in 2022. Perhaps what we see here is just an extension of that.could they in theory do the same to a human brain
Mr Musk is talking about carrying people's brains around on a chip so it's all starting to come together. I've met people whose brain would fit on a floppy disk.
And talking of D. melanogaster, time flies like the wind but fruit flies like a banana.
Gary C said:
Thats the real question isn't it
could they in theory do the same to a human brain
Yes off course but you have billions to map first then build up the cogntive and system functions.could they in theory do the same to a human brain
Human brains are not much different from other animals just possess more cognitive processing albities.
Flyingakite said:
Yes off course but you have billions to map first then build up the cognitive and system functions.
Human brains are not much different from other animals just possess more cognitive processing albities.
Of course and its unrealistic, but it seems, theoretically possible.Human brains are not much different from other animals just possess more cognitive processing albities.
Angstrom by angstrom slicing and mapping machine, lovely
It s not just the complexity of the neurons and connections between them, it s also the transient, dynamic state of them all as well.
I think we ll get to a point where it can be all modelled, but I don t believe we ll ever be able to copy a person entirely.
Basically, it s like a road network. You can have all the same roads, you can copy the traffic light sequencing even, but you d also need to copy the exact state of every single vehicle on that road, instantaneously. I reckon that will be the hard bit. And then on top of that, identically mirroring all the inputs, also hard.
But I do think we ll be able to produce an artificial intelligence that mirrors many of our own capabilities. The Era Of Experience by David Silver is quite interesting to read on this topic.
I think we ll get to a point where it can be all modelled, but I don t believe we ll ever be able to copy a person entirely.
Basically, it s like a road network. You can have all the same roads, you can copy the traffic light sequencing even, but you d also need to copy the exact state of every single vehicle on that road, instantaneously. I reckon that will be the hard bit. And then on top of that, identically mirroring all the inputs, also hard.
But I do think we ll be able to produce an artificial intelligence that mirrors many of our own capabilities. The Era Of Experience by David Silver is quite interesting to read on this topic.
Edited by Arnold Cunningham on Sunday 31st May 13:07
Gary C said:
Flyingakite said:
Yes off course but you have billions to map first then build up the cognitive and system functions.
Human brains are not much different from other animals just possess more cognitive processing albities.
Of course and its unrealistic, but it seems, theoretically possible.Human brains are not much different from other animals just possess more cognitive processing albities.
Angstrom by angstrom slicing and mapping machine, lovely
Gary C said:
Flyingakite said:
Yes off course but you have billions to map first then build up the cognitive and system functions.
Human brains are not much different from other animals just possess more cognitive processing albities.
Of course and its unrealistic, but it seems, theoretically possible.Human brains are not much different from other animals just possess more cognitive processing albities.
Angstrom by angstrom slicing and mapping machine, lovely
Arnold Cunningham said:
It s not just the complexity of the neurons and connections between them, it s also the transient, dynamic state of them all as well.
I think we ll get to a point where it can be all modelled, but I don t believe we ll ever be able to copy a person entirely.
Basically, it s like a road network. You can have all the same roads, you can copy the traffic light sequencing even, but you d also need to copy the exact state of every single vehicle on that road, instantaneously. I reckon that will be the hard bit. And then on top of that, identically mirroring all the inputs, also hard.
But I do think we ll be able to produce an artificial intelligence that mirrors many of our own capabilities. The Era Of Experience by David Silver is quite interesting to read on this topic.
Ah, are you saying they basically copied the architecture and ROM of the fruit fly but for a human, we would need to copy the RAM state too ? I have no idea how 'memory' is stored in a person.I think we ll get to a point where it can be all modelled, but I don t believe we ll ever be able to copy a person entirely.
Basically, it s like a road network. You can have all the same roads, you can copy the traffic light sequencing even, but you d also need to copy the exact state of every single vehicle on that road, instantaneously. I reckon that will be the hard bit. And then on top of that, identically mirroring all the inputs, also hard.
But I do think we ll be able to produce an artificial intelligence that mirrors many of our own capabilities. The Era Of Experience by David Silver is quite interesting to read on this topic.
Edited by Arnold Cunningham on Sunday 31st May 13:07
There are animals with more nuerons in their brain than humans.
The question is what defines humans as unique when in reality the brain evolved over about 10 million years from a common ancestor. So 10 million years to make the modern brain , isn't difficult really at the pace of technology.
The question is what defines humans as unique when in reality the brain evolved over about 10 million years from a common ancestor. So 10 million years to make the modern brain , isn't difficult really at the pace of technology.
Gary C said:
Ah, are you saying they basically copied the architecture and ROM of the fruit fly but for a human, we would need to copy the RAM state too ? I have no idea how 'memory' is stored in a person.
They emulated a generic, vanilla fruit fly, not this particular fruit fly who is called Dave and particularly enjoys Russian literature, classical Japanese artwork, and rotten peaches. It’s showing hardwired instinctive behaviour. They did the same thing a few years ago with the simpler connectome of a nematode worm, and found the same thing. In fact if I remember rightly they also put it into a robot.
Gassing Station | Science! | Top of Page | What's New | My Stuff