🚨140,000 NEURONS. 50 MILLION CONNECTIONS. ZERO TRAINING. THEY PUT A BRAIN INSIDE A COMPUTER AND IT MOVED.
The video shows a simulated fruit fly controlled by a digital copy of its biological neural wiring.
No reward model. No giant dataset. No billion-dollar training run. The behavior comes from reconstructing the connections that already produced intelligence in nature.
That is what makes Google’s connectomics work bigger than another brain visualization. The map is becoming executable.
We went from tracing 302 neurons in a worm by hand to processing 1.4 petabytes of human brain tissue with AI. Every new map exposes architectures evolution spent millions of years testing.
Current AI tries to learn intelligence from outputs. Connectomics gives researchers the wiring underneath them.
The next breakthrough may not come from scaling an artificial brain. It may come from finally reading a real one.
ONE OF CLAUDE'S DEVELOPERS TOLD ME:
"YOU'RE USING AI AT 5%."
I expected some crazy setup.
It was just a folder with text files.
Then he opened the graph.
Thousands of notes. Everything connected.
"I didn't link any of this. Claude did."
Then he typed:
where am i contradicting myself?
10 seconds later it returned old notes, dates, and quoted his own words back to him.
He laughed:
"No friend will ever tell you this. But this thing will."
That changed how I think about AI.
Most people use it like a chatbot.
A few use it like a mirror.