Could a human brain run on a computer?

In a theoretical sense, maybe—but not with anything we can do today.

If by “run on a computer” you mean digitally copying or simulating a human brain, then it may be possible in principle. But that remains unproven, and there are major scientific and philosophical hurdles.

The main obstacles are:

1. We cannot map a whole brain at the needed detail yet. A human brain has about 86 billion neurons and on the order of 100 trillion synapses, and we do not currently have a non-destructive way to capture all of that at the resolution needed for a faithful simulation.

2. A brain is more than wiring. A realistic model would likely need to include synaptic plasticity, chemical signaling, glial cells, and other biological processes, not just neuron-to-neuron connections.

3. The computing burden would be enormous. Even if we had a perfect scan, simulating a brain in enough detail to reproduce its behavior in real time would require far more memory and processing than typical computers can provide, though exact requirements are still uncertain.

There is also a deeper question: even if a computer could reproduce the brain’s behavior perfectly, would that be the same mind, or just a copy? We do not know.

So the short answer is: possibly in principle, but not yet in practice, and it remains scientifically and philosophically unresolved.