TL;DR: We can solve the problem of preserving consciousness during mind uploading by preserving my living biological neurons and their connectome in a spatially distributed network spanning hundreds of kilometers. Individual neurons can be later cloned or repaired.
There is a problem: we don’t know if digital uploads will be truly conscious, and even if they are conscious, will they have my consciousness? However, we want uploading as it is the only long-term solution for the immortality problem, because uploads can be copied, backed up, and repaired. Some people argue that transhumanists will “kill everyone” by replacing living humans with digital p-zombies and thus their desire for immortality is deadly. While we can address these concerns via philosophical arguments about functionalism, they will never be sufficient.
Here I suggest a way to escape the problem by preserving living biological neurons as supposed carriers of consciousness, but placed far from each other and connected via digital means instead of axons. The procedure in a nutshell is the following:
Using advanced nanotech, we take each neuron (or group of neurons) from a living human brain and place it at a great distance from other neurons, so the whole brain is now distributed over a 100 by 100 km space (there is a limit on the size of the mesh imposed by the speed of light, so the brain as a whole will be able to work at the same speed as a human brain).
Each neuron continues to live and fire the whole time.
Each neuron is now placed on a separate digital chip.
Each neuron is connected with all other neurons via digital communication lines which send information about the firing of one neuron to other neurons in exactly the same way as it was done by the human connectome before. Axons are replaced with digital transmitters at some points.
From an external point of view, it is the same human brain but just distributed over a large distance.
From an internal point of view, nothing changes.
If we think that human consciousness depends on neurons, we are sure that the brain is conscious.
Now we can clone this distributed brain by adding more biological neurons (from the same person—evolved from their stem cells) into the network in parallel and synchronizing such neurons with original neurons. After that we can replace half of the neurons in the person with the new ones and use the other half for a copy. Both copies get the same amount of the original neurons.
After we can create biological copies and we can store some copies “frozen” as an archive in case of catastrophes.
The distributed biological brain is almost indestructible as any neuron can be cured or replaced, and only a large meteorite strike or electricity collapse can kill all of them.
If the whole scheme seems extremely expensive but philosophically sound, we can find ways to make it cheaper. One way is to distribute large groups of neurons like cortical columns. Anyway, the whole scheme assumes advanced nanotech.
Single neurons on a chip and brain organoids-on-a-chip already exist. Cortical Labs released CL1, which is a biocomputer using 800,000 lab-grown human neurons interfaced on a silicon chip. Each chip costs $35,000, with a cloud version at $300 per week.
Counterarguments:
1 Consciousness depends not on neurons and synapses, but on something which is destroyed via such “brain distribution:”
A) Global quantum effect is disrupted. Some may say that morphological quantum fields between neurons – or whatever quantum woo you have except microtubules, as they will be mostly in place in neurons – will change, but if it is measurable, we will emulate it too.
B) Global electromagnetic effect is disrupted. Neurons affect each other not only via axons but via global electromagnetic fields.
C) Axons are needed for consciousness. Also, one may argue that consciousness is inside axons but not neurons, or that axons are sending something more than just information about spikes, like spin or some other quantum property.
We assume that some simpler ways of communication between neurons like neurotransmitter gradients, direct connections, blood vessels, glia, blood sugar, and oxygenation are also emulated together with spike potential.
In other words, the main assumption is that the whole complexity is inside neurons, but not in the gaps between them. This allows us to “cut” the brain into high-complexity regions with low-complexity communication between them. If there is no decline of complexity in the gaps between neurons, the whole idea is doomed.
2 There will still be doubts about the sameness and consciousness, similar to uploading. While such doubts will be smaller, the price of biouploading will be astronomical. A project like this could easily cost hundreds of billions of dollars per person. Even if nanotech makes it very cheap, it will still take a lot of land, which has a price. In other words, we pay a lot but do not solve our philosophical doubts. Most people will have to choose between cheap uploading or even cheaper LLM-sideloading – or death.
However, if we can solve uploading once, we can gain philosophical insights and also limit our uncertainty about the true carrier of consciousness.
Two Similar Ideas:
Another similar idea is biouploading-to-blank-brain-organoid: we take a “blank” brain – maybe a brain specially grown for that from my stem cells, but without any experiences (it should be regarded as a large organoid) - and copy my connectome structure and synaptic weights into it.
Biouploading into “artificial neurons” - here we do not use real biological neurons but use instead analog electric models of living neurons, which are small condensers with some internal wiring. In some sense they are similar to vacuum tubes, which are famous for their warm sound.
Biouploading: Preserving My Living Neurons and Connectome as a Spatially Distributed Mesh
TL;DR: We can solve the problem of preserving consciousness during mind uploading by preserving my living biological neurons and their connectome in a spatially distributed network spanning hundreds of kilometers. Individual neurons can be later cloned or repaired.
There is a problem: we don’t know if digital uploads will be truly conscious, and even if they are conscious, will they have my consciousness? However, we want uploading as it is the only long-term solution for the immortality problem, because uploads can be copied, backed up, and repaired. Some people argue that transhumanists will “kill everyone” by replacing living humans with digital p-zombies and thus their desire for immortality is deadly. While we can address these concerns via philosophical arguments about functionalism, they will never be sufficient.
Here I suggest a way to escape the problem by preserving living biological neurons as supposed carriers of consciousness, but placed far from each other and connected via digital means instead of axons. The procedure in a nutshell is the following:
Using advanced nanotech, we take each neuron (or group of neurons) from a living human brain and place it at a great distance from other neurons, so the whole brain is now distributed over a 100 by 100 km space (there is a limit on the size of the mesh imposed by the speed of light, so the brain as a whole will be able to work at the same speed as a human brain).
Each neuron continues to live and fire the whole time.
Each neuron is now placed on a separate digital chip.
Each neuron is connected with all other neurons via digital communication lines which send information about the firing of one neuron to other neurons in exactly the same way as it was done by the human connectome before. Axons are replaced with digital transmitters at some points.
From an external point of view, it is the same human brain but just distributed over a large distance.
From an internal point of view, nothing changes.
If we think that human consciousness depends on neurons, we are sure that the brain is conscious.
Now we can clone this distributed brain by adding more biological neurons (from the same person—evolved from their stem cells) into the network in parallel and synchronizing such neurons with original neurons. After that we can replace half of the neurons in the person with the new ones and use the other half for a copy. Both copies get the same amount of the original neurons.
After we can create biological copies and we can store some copies “frozen” as an archive in case of catastrophes.
The distributed biological brain is almost indestructible as any neuron can be cured or replaced, and only a large meteorite strike or electricity collapse can kill all of them.
If the whole scheme seems extremely expensive but philosophically sound, we can find ways to make it cheaper. One way is to distribute large groups of neurons like cortical columns. Anyway, the whole scheme assumes advanced nanotech.
Single neurons on a chip and brain organoids-on-a-chip already exist. Cortical Labs released CL1, which is a biocomputer using 800,000 lab-grown human neurons interfaced on a silicon chip. Each chip costs $35,000, with a cloud version at $300 per week.
Counterarguments:
1 Consciousness depends not on neurons and synapses, but on something which is destroyed via such “brain distribution:”
A) Global quantum effect is disrupted. Some may say that morphological quantum fields between neurons – or whatever quantum woo you have except microtubules, as they will be mostly in place in neurons – will change, but if it is measurable, we will emulate it too.
B) Global electromagnetic effect is disrupted. Neurons affect each other not only via axons but via global electromagnetic fields.
C) Axons are needed for consciousness. Also, one may argue that consciousness is inside axons but not neurons, or that axons are sending something more than just information about spikes, like spin or some other quantum property.
We assume that some simpler ways of communication between neurons like neurotransmitter gradients, direct connections, blood vessels, glia, blood sugar, and oxygenation are also emulated together with spike potential.
In other words, the main assumption is that the whole complexity is inside neurons, but not in the gaps between them. This allows us to “cut” the brain into high-complexity regions with low-complexity communication between them. If there is no decline of complexity in the gaps between neurons, the whole idea is doomed.
2 There will still be doubts about the sameness and consciousness, similar to uploading. While such doubts will be smaller, the price of biouploading will be astronomical. A project like this could easily cost hundreds of billions of dollars per person. Even if nanotech makes it very cheap, it will still take a lot of land, which has a price. In other words, we pay a lot but do not solve our philosophical doubts. Most people will have to choose between cheap uploading or even cheaper LLM-sideloading – or death.
However, if we can solve uploading once, we can gain philosophical insights and also limit our uncertainty about the true carrier of consciousness.
Two Similar Ideas:
Another similar idea is biouploading-to-blank-brain-organoid: we take a “blank” brain – maybe a brain specially grown for that from my stem cells, but without any experiences (it should be regarded as a large organoid) - and copy my connectome structure and synaptic weights into it.
Biouploading into “artificial neurons” - here we do not use real biological neurons but use instead analog electric models of living neurons, which are small condensers with some internal wiring. In some sense they are similar to vacuum tubes, which are famous for their warm sound.