A worm with 309 neurons. Mapped completely since 1986. Forty years later, no simulation reproduces its behavior. Joscha Bach (@Plinz) on what that means for BigTech's plan to upload a human brain. youtu.be/CzjWGkXlK8k
🤖 Made with AI

May 9, 2026 · 1:06 PM UTC

57
31
8
244
48,340
RelevantRecentLikes
I wish I could figure a way to make money from people willing to "upload" their brains. Maybe sell them virus insurance. So that if when they are in the Great AI Beyond, they can be protected from computer viruses which would turn their heavens into hells.
3
16
656
You can build an antenna in software but it won’t pick up a signal…
1
2
300
the 309-neuron sim gap is a grain-of-detail failure. wiring diagram is topology. each neuron has thousands of tunable internals: microtubule states, neurotransmitter mix, ion channel dynamics. mapping the diagram doesn't capture the cell.
23
778
"...what that means for BigTech's plan to upload a human brain..." Nobody should treat that idea seriously.
3
1
19
602
Having a map of a neuronal representation is like having a database with all of the lines but none of the data.
1
6
578
But didn't the upload a fruitfly recently, which is currently humming away in someone's PC?
2
4
423
@alexwg's work on the entire uploaded/simulated fruit fly connectome may interest you. Its behaviour is ably reproduced: nitter.cf/pascal_bornet/status/2…
Most AI companies are trying to simulate intelligence. Meet Dr. Alex Wissner-Gross, co-founder of Eon Systems, a startup trying to copy a real brain and run it inside a machine. That is why this story is so hard to ignore. Instead of building another artificial model, Eon took a very different path: copy the biological wiring itself, connect it to a body, and see what happens. And something did happen. The digital fly walked. It groomed itself. It foraged for food. No training. No programming. No machine learning. That is the part that really stayed with me. For years, the field kept splitting the problem in half: → a brain with no body → or a body driven by artificial training Eon tried to close that gap. It copied a fruit fly brain neuron by neuron, attached it to a simulated body, and turned it on. The behavior came from the wiring itself. That is not the usual AI story. And it gets even bigger from here. The fly was the proof of concept. Now the company is going after the mouse brain, vastly more complex, and eventually humans. That is why this feels less like a demo and more like a warning shot. What do you think this really is: a new path to AI, or the first step toward digital minds built from biological ones? #AI #ArtificialIntelligence #Neuroscience #BrainEmulation #DeepTech #Innovation #Technology #FutureOfWork #Robotics
1
7
849
The brain is fundamentally a hybrid analog-digital system, not purely one or the other. Both neurons and synapses exhibit a mix of analog and digital. Dendritic computation is often highly analog and nonlinear.
1
404
WAHT!? YOU INTERVIEWED MR. BACH!? NO WAYYYY! Be careful, Mr. Keating! He looks favourable on Expanding Earth Theory and he might even disagree with other mainstream stuff!
1
1
281
Neurons without an entropy driver are just mechanical systems like any other.
1
240
On LessWrong someone gave a great explanation: Connectomes are the biological equivalent of what you'd get if you removed all the component symbols from a circuit schematic and left only the wires. Good luck trying to reproduce the original functionality from that data.
1
2
598
nitter.cf/lilithdatura/status/18… I dropped some of this information into spaces in January for a tangent conversation, I flagged it as being very important for multiple reasons. It wasn’t your spaces though. @TheVaultMother
Replying to @ClearOrbit
Same here, I fleshed out some thoughts with Grok on it (in my most recent post), and basically I'm on the fence. Grok helped reinforce my understanding of the feedback loops this 4Chan post was referring to (not sure who really put this out tho), as well as the LNN (Liquid Neural Networks) that would make this tech possible, something I learned about a few years ago, oddly enough from the C. elegans (worm).
1
1
1
310
Worms have a distributed brain.
17
you've missed the functionality of neurons, like calling a postcard a movie
108
I don't even know what this is about. Too distracted by OP's jawline.
29
There's a ROM chip somewhere inside them to boot them up. We haven't found it yet, huh :(
58
Every tree in the forest has been mapped and no understanding of the ecosystem has emerged. DNA has been sequenced and the code seems incomplete. The structure of a dead cell and a living one is very similar, yet ...
2
2
339
The more I learn about the world the more it seems like this whole place is a pyramid scheme-like path system for ever more complex forms of consciousness
1
2
131
You can probably get somewhere using AI, but: Did you learn anything? Is the AI now a nemetode?
1
22
No wet Brain, no consciousness or Soul. A complete Brain map can only be done by the Great fine tuner.
1
70
40 years a product of evolution has tried to mimic the effects of billions of years of evolution. and you’re surprised ? ur a professor of who exactly ?
1
87
Ever been fishing in a map lake? Or walk down a map street? Or climbed a map mountain? Ever confuse a description of your home with your actual home? Or taken a nap in a description of a warm comfy bed? Felt the texture of a tree while using an representation of it as a model?
1
1
1
151
0 neurons here, good luck to train a model for this
1
71
What about the Open Worm Project?
1
162
Yeah they have done this actually: openworm.org
1
298
Well, you probably need to map the rest of the cells, right?
1
1
172
I don't believe they know what causes consciousness to emerge. All that money and time...
1
1
380
I think it is because it is not connected to the worm's own purpose built sensors that's connected to these neurons. Like a lawnmower connected to a battery charger minus the battery itself would cause the system to not behave as its supposed to. The battery closed the circuit.
1
888
Los aviones vuelan, pero yo no les veo mover las alas... Lo que hay que capturar y reproducir es la función, no la forma. La forma está a años luz de ser descifrada, y si no que pregunten a Blue-brain y Human-Brain projects.
2
only a part of it this little bio network is much like an LLM is today, without any kind of loop of perception/sensation through the "tools" of its body, it's like taking a single frame from a video and claiming "where is the movie here?"
31
Life cannot be simulated. It is an open dynamic system whose phase space is changing. Read my book: Disrupt Science. The Future Matters. At least try to understand the subject before producing non-sensical statements!
13
Anyone with full interview?
17