Could you upload your mind to live forever?

Current brain scanning techniques unfortunately destroy the brain tissue and cannot be used on a living organism. There is some work on developing brain preservation techniques to make future brain scanning easier. Since the brain might be leveraging quantum effects for thought, we might have to lean on quantum mechanics to fully scan or understand or emulate what's happening in a brain.

Although brain scanning is a very hypothetical technology, we can see a potential path to success. At the outside, the advent of the singularity is likely to provide mind uploading technology. We discussed the connection between brain scanning and brain computer interfaces at the end of the video.

#mindupload #bci #immortality

Can You Upload Your Mind & Live Forever?

Digital immortality: would you upload your mind to a computer?

A startup is pitching a mind-uploading service that is “100 percent fatal”

Scale of the Human Brain – AI Impacts

0:00 Intro
0:19 Contents
0:25 Part 1: What is mind uploading?
0:49 Why uploading might be a good idea
1:23 Digital intelligence and the singularity
2:04 Is the new body you, or more like a clone?
2:30 Two main steps to mind uploading
2:54 Part 2: Brain scanning
3:19 Exploration of synapses
4:00 From teraflops to exaflops (1,000,000x)
4:43 Current brain scanning techniques
5:36 We need to learn to read the book
5:59 Book recommendation: House of Suns
6:21 Example: Nectome preservation startup
7:21 Stochastic brain scanning approaches
8:12 Quantum simulation approaches
9:01 Rather crazy quantum ideas
9:29 Part 3: Running mindware
9:43 Example: emulating C. elegans
10:22 Virtual brain properties
10:50 Cloning virtual brains (no secrets)
11:27 Timelines for mind upload
12:07 High-bandwidth brain computer interfaces
12:49 Conclusion

[ct]

Philosophical: My take is, we are essentially already copies of a past state of us. The me right now is already not the me of 5 years ago ( that is the state of consciousness. Jeff Hawkins from Numenta also has an interesting take on consciousness as such. I think the emphasis on current state is very important instead of just naming it consciousness like a static thing that exists over all times). The atoms are entirely changed via metabolism (my past me in terms of the atoms is now completely distributed to the environment). Only the pattern in which atoms are assembled, specifically the patterns of my neurons forming my memories, are somewhat consistent There are a lot of though experiments like the ship of Theseus or splitting all my matter of me in two versions and fill the empty spaces with new atoms, who is the original? The only reason why I think of myself as the me of 5 years ago is my access to some portion of these memories (that are not accurate, but then they are the only reference that I have). People can loose this access (amnesia for instance) and don’t associate themselves with their past states any more. These patterns, the memories are all that matters. A copy with this access would be no less a continuation of a past state then if the biological instance remains. None would be more the me that exists now, than I am the me that did exist 5 years ago, it’s already an illusion. If someone deleted my memories formed today and shoed me a video of me going theough the day, I would say it is a fake and wouldn’t associate myself with this individual. Still in that deleted timeframe this me would have been conscious, the memories just didn’t make it. If the biological components of one’s mind would slowly replaced by non-biological components or just by parallel running backups on a computer and the biological components are slowly dying of, we wouldn’t even notice (on a biological level this already the case). A rapid or a discontinuous transition to a purely non-biological thinking substrate would actually not be different; the gradual one could merely allow us our illusion of a continuous self. There wouldn’t be a biological future me that would have otherwise existed, which in any case wouldn’t have been the past state. Instead there would be a virtual future me, that also wouldn’t be the past state. One could also keep the biological self with extended neocortex (like a biological platform to access of the greater me-collective) and keep it synchronized with the rest (see below). One might also view it like this, there are a large amount of potential future me’s (versions what might happen, potential states of consciousness that retain memories from my present state). I don’t really care which of them will be the one to come into existence, within curtain constraints of course (not the one in rokos basilisk ;-) as long as it is one of them (so there is a future me, or many mes at different points in time, but I don’t know which). In a many-worlds interpretation of quantum-physics potentially all of them, potentially also in the technical solution.

Technical: The idea is to have both a high bandwidth decentralized connection via nanobots (or more advanced neuralinks … the point is to have a high bandwidth bidirectional connection between neurons and the cloud) and one or more digital backups. So a biological instance that remains plus potentially copies (potentially even other biological ones). But they would not exist like mere copies. We could hybridize the biological instance/es by extended artificial neocortex in the cloud, which is luckily already modular thanks to cortical columns (this might also allows us to become smarter, the idea comes from Kurzweil). In the nonbiological portion we might even apply an equivalent of backpropagation to analytically adapt the synaptic connections to instantly learn things or change personality patterns at will etc. Moreover, the other purely digital instances (that consist of a purely digital backup of both the biological backup plus its artificial neocortex) could go on living parallel lives. But similar to current AI models, anything the instances learn would be synchronized (perhaps just via the artificial neocortex portion, which might quickly be dominant relative to the biological one). Or better yet, there could be a common memory pool and synchronization might be done according to the instances specific specs. Synchronization of memories of various instances having monogamous relationships with different people might otherwise be problematic to name one example. But then, one could essentially take all of lives possible paths without compromising by not going others. Instances could merge and split on demand (selective synchronization on or off, specific memories on or off). Most instances would be fully digital, living in realistic VR, with others having physical embodiments, biological or otherwise, which could also act more as physical platforms for the different minds. That way there would be no old biological me that would essentially die in a destructive process. Simultaneously, the biological or hybridised instance (or multiple of them) would still have all the advantages of mind uploading and access the all the experiences. As a scenario, a biological instance goes to sleep or something similar (no consciousness, also the non-biological portion of this hybridized mind would be dormant) and a purely virtual backup with recent access to that state would go on doing stuff in the cloud. Internet access might even be cut for a while, so there is no connection). Later on the information is synchronized again with the biological instance (or just the artificial neocortex portion of it). The digital backup might lie dormant again at that point. From the subjective experience of the biological instance, waking up in the morning, it would be like having left the biological body entirely (without any connection at that point) and jumped back into it later on. From his memory, while being a disembodied entity in the cloud he would remember that at that point he would not have cared (provided there are also physical backup bodies) if in the offline-time something would have happened to the biological body.

[bigcauc]

I don't think consciousness is transferable in a literal sense of moving from one to another. I think it may very well be possible to copy and stimulate said conscious experience with sufficiently advanced technology and understanding, but once you die, YOU die. The only thing I can imagine working is keeping the brain alive and active in a robotic body that can simulate a nervous system and all of the senses. I think the brain is adaptable to a degree.

[MrNicogp]

Excellent!!! And a very good segue for a piece on Penrose-Hameroff and (claimed) non-computability of "understanding" and consciousness;)

[vernongrant]

Once we have a computer a billion times smarter than a human, mind uploading will be possible. That will probably be at the dawn of the Singularity.

[MichaelDeeringMHC]

Another book to order. Thanks for the recommendations. I'm currently finishing book 2 of the bobiverse, and 3 and 4 are sitting on my desk.

[DanHammonds]

I love the idea of being uploaded to a heavenly digital universe where anything goes and I have no physical restrictions or pain, but I'd be too scared that some sadistic engineer or rogue AI would use it to torture us for eternity and we would never be able to communicate with the outside world to call for help.

[danielchristiansen]

I think the biggest issue here is that we don't understand what consciousness is -- can't even define it. Maybe once we start augmenting human brains with external enhancements, we'll see that consciousness can "jump" to whatever hardware it is running on. But maybe instead, perceived consciousness is tied somehow to some biological mechanism that we don't really understand yet. But if and when we did, maybe that too could be emulated.

[erobusblack]

Mind upload requires a brain computer interface using EEG and infrared As well as FMRI but that may not be required later it might just be an EEG infrared headsets, Cross-referencing that data from real-time mixed reality data and training a virtual human on imitation learning admittedly the mixed reality and virtual human andThe brain scanner all have to be up to par to do the task and it takes pretty much the rest of your life to achieve an upload. Also want to point out that a lot of the functions going on in the brain are things that a virtual human aren't going to need like a heartbeat and lungs, I want to point out the experiments where they're using AI and brain scanning technology to literally read minds. Using weak AI models like GPT2 and stable diffusion they're capable of collecting recreations of images shown to the person and what the person is thinking, So with the right model you could cross-reference similar types of data with what a person is doing in mixed reality which is Real-time data of what the person's doing and use a virtual human that has the ability for object interaction and computer vision, As well as all the other 5 senses and the ability to move around in the environment a virtual environment of course. And over enough time and asking the right questionnaires to collect recreations of memories, It would be a Theseus ship of course but it would be a mind upload, Also consciousness is not as profound as people want to make it out to be. It's just the story you tell yourself about the world the story you tell yourself about yourself and the story you tell yourself about yourself within that world, And does require memory and persistent experience. People often mix up the word consciousness with the word cognition a combination of different cognitive functions such as and self-awareness and other types of memory and forethought And so forth

[Me__Myself__and__I]

2045 for the singularity is probably pessimistic. Humans really suck at estimating things that are exponential. Assuming there are not enforced international laws slowing things down, the singularity will likely take between 6 months to 5 years after AGI becomes a reality. So the key question is when does AGI happen? It is probable that it will happen in less than 20 years given current knowledge and projections (like transformers continuing to scale). AGI is fuzzy and will likely be emergent, so no one can be certain. It could happen as early as 2024 and probably will happen sometime this decade considering the current state of AI (a cash fueled race with little to no concern for safety). So reasonably good chance the singularity will occur somewhere between mid 2024 to 2034 given current conditions and factoring in exponentials. Sooner probably means more dangerous.

[Citrusautomaton]

When ASI comes along i feel like Mind-uploading won’t be far behind. Hope so at least, can’t exactly trust my weak cardiovascular system to go on for more than 20 years.😳

[jverart]

Ghost in the shell is being left behind with all this craziness. Thanks for the book recommendations too!

[ujccchp]

I think we need a 3-dimensional map of the physical brain first, bit by bit. X load it

[freedom_aint_free]

Consciousness is a illusion caused by the integration of the various modules in our brain, the consciousness is the illusion of self, is the coherence of the "operational system" plus sensors and hardware. In principle is perfect possible to copy your "consciousness" to wherever can run the code or emulate the original hardware, but it would not be you as there's no you, only instances of a plan, architecture plus life experiences or the history of the system's usage. It's like cars: There are Toyotas Camrys 2023, but no single one car of there brand and year is the Toyota Camry, for it's a concept a abstraction.
In the case of mind upload it becomes obvious when you consider that once you have your mind digitized you can transfer it more than once to more than one hardware at a time, and so those copies will be different instances of you and once they start having different experiences they will diverge and stop becoming you the second they get uploaded.

[etunimenisukunimeni]

The way I understand it, there's more to mind uploading than very good/perfect scanning technology. You also need to know what's happening in the synapses - there are neurotransmitters flowing in and out and god knows what else, all of which is an integral part of the functioning brain. You can catalogue every cell and its connections in the brain and still end up not knowing how it all works. Don't get me wrong, having a perfect scan helps A LOT, but it just might not be enough.

[epg-]

If the process can be done gradually while the patient is awake, it would be possible to test the copy/cut problem. For example, you could transfer the entire visual cortex, then ask the patient if they see what's being presented to the simulated brain.

[cours]

problem is that we currently can't model the brain, we don't know why neurons makes connection to other neurons for example so even with a perfect scan an enought computation power, we won't be able to "run it".

[BCIQW]

If you haven't already, you may be interested in looking into Duke university's new brain imaging technique which is 64 million times more detailed than MRIs.

[dornwilliamson]

Hello, I am interested in the brain 🧠 I had a severe TRAUMATIC BRAIN INJURY from a Head On Car Crash Aug 29th 1987 I lost some brain cells, My left side of my brain was bruised and my brain stem was twisted. I was in a state of coma for about four months.
Once I was awake I had to relearn everything.
I lost my sense of taste and my sense of smell. I lost the right field Of vision from of both eyes.
Do you have any ideas Of how to repair my brain cells that were lost?!

[noahlovotti]

We need to work more on longevity research first.

[perfectscotty]

Kurzweil has talked about how their isn’t much research going into mind uploading. I think it’s the only way to go.