Key Takeaways
- Hodak rejects digital snapshot uploads: a software replica matching your behavior fails if it lacks continuous subjective experience.
- Science Corporation showed direct neural compute works in humans with PRIMA, a retinal implant that gained CE mark approval in Europe by restoring functional form vision.
- People tolerate massive identity changes over a lifetime, but only if they experience uninterrupted phenomenal continuity.
- Full connectome mapping is the primary blocker for brain emulation, and Hodak states the field is now close to executing the project.
The Static Copy Fallacy
Silicon Valley often treats mind uploading like cloning a hard drive. You scan the brain, write a software model, and boot it on a server. Max Hodak thinks that logic falls apart the moment you test the subjective experience.
“If I just scanned your brain into a computer and there is a software simulation of you, does that count as you?” Hodak asked Guo. “Would that make you feel better about dying of cancer?”
For most people, the answer is no. If a clone wakes up in a data center while your biological body dies in a hospital bed, your subjective thread ended. You did not escape mortality; you built an actor that impersonates you after you die. A model that mimics your tone or matches your decisions does not solve the problem of personal survival.
Continuity Over Output
Human identity handles drift surprisingly well. Your memories, physical cells, and opinions at age thirty share very little with your state at age seven. You accept that drift because the transition was smooth.
“I'm in the camp that continuity is greatly important,” Hodak explained. “People will accept significant drift in their identity over time as long as they have continuity.”
This distinction exposes the flaw in modern AI comparisons. Large models mimic human answers, but they do not maintain a live, continuous state. Hodak pointed out that preserving identity requires unbroken phenomenal flow: “If you have a software simulation that answers exactly like you would now, but it's not phenomenally continuous, that is less satisfying.”
Connectomics and Substrate Independence
Science Corporation focuses on treating the nervous system as a computational engine rather than a purely chemical mystery. Their PRIMA retinal implant restored functional form vision to blind patients and earned European CE mark approval by treating optical input as hardware signals.
From that foundation, Hodak sees a path toward moving human minds off fragile biological tissue. The goal is substrate independence: transferring consciousness into non-biological hardware without snapping the thread of experience.
“One of the big missing pieces here is connectomics,” Hodak noted. “That is getting pretty close, I think. In terms of to the point where the project could be done.”
Once researchers can map the full connectome and interface with it directly, you can replace biological circuits piece by piece. When the transition happens gradually, continuity stays intact. You do not die and leave behind a digital ghost; you move into a durable substrate without ever going dark.
What to Do With This
If you build systems that model user state, stop relying on discrete snapshot saves that destroy context. Audit your core architecture this week: track whether your data pipelines preserve continuous state history or just overwrite records with static behavioral outputs.