Biology is too Hard

[jordansparks]

To revive someone biologically, we would need nanotechnology, the ability to manipulate individual atoms. We are a very long way from having any sort of technology like that. In contrast, the technology needed to upload someone as a Whole Brain Emulation (WBE) is much more simple. We must basically be able to see the outlines of cells on scans, along with some molecular tags. The difference in size between an atom and a cell outline is, let's say, 100. This means that biological revival requires a scanning resolution of about 100 times smaller than uploading. Unfortunately, biological revival also requires robotic manipulators at that smaller scale. Uploading only requires scanning at the larger resolution, not the additional requirement of being able to build actual machines at that scale. Building small machines is very hard. We could say it's 100 times harder than visualizing the same structures in a microscope. 100 x 100 = 10,000. This makes biological revival at least 10,000 times more complex than uploading, probably a lot more. This is obviously a qualitative number rather than an accurate quantitative number.

When something is more than 10,000 times easier, it's going to get done much sooner. Simple math tells us that uploading will be available much sooner than biological revival. Nobody will care about biological revival. You might think you want it right now, but it won't really be an available option in any sort of reasonable time frame. That's a big gap between what you want and what you can have. You'll settle for you what you can have. Everyone will upload.

[Wesxdz]

Have you considered running some sort of Kaggle competition on computer vision for turning a tensor of brain voxels into a graph data structure?
Is the primary bottleneck now mainly robotic manipulators for slicing and dicing and synapse-voy... flood filling networks, or computational reconstruction of the brain?

https://claude.ai/share/2f95b9ae-1428-4 ... 0dcc74a5c0
(fixed to public link)

[jordansparks]

The current bottleneck is mostly tracing the wiring. That step would cost about $10 billion right now for a single mouse. We need the software to get a lot better. Slicing and dicing certainly isn't cheap either, but nowhere near as expensive as tracing. Once we get better at tracing, there will be a series of additional bottlenecks.

[Wesxdz]

Algernon is truly getting the C-Suite experience.

[Mati_Roy]

epistemic status: poetic

imagine being 1 in a million in terms of predicting the future, but you specify in your cryonics contract you want a biological revival, so by the time you're revived you're multiple eras behind and your mind upload friends have become unrecognizable hive mind Gods who've lived a thousand lives, and instantly convince you to upload anyway but had to wait to respect your wishes and ask for your consent because you didn't leave that decision to our wiser descendants

https://x.com/matiroy/status/2098831033382932841

[jordansparks]

True, but my opinion has changed slightly. Since I now think uploading will be low tech, I don't think the people performing the uploads will be particularly wise. I used to think they would be very advanced, but I now think they will just be ordinary people like us.