The Quest to Build a Brain: Why 2023 Came and Went

In the mid-2000s, neuroscientist Henry Markram launched one of the most ambitious projects in scientific history: digitally reconstructing the human brain, neuron by neuron, synapse by synapse. The Blue Brain Project, and later the European Union's billion-euro Human Brain Project, promised something extraordinary—a complete cellular simulation of a human brain by 2023. Spoiler alert: we're past 2023, and your brain hasn't been uploaded to the cloud yet.

So what happened? Were these projects failures, or did scientists simply underestimate the most complex object in the known universe?

The Reverse-Engineering Approach

The methodology was sound, even ingenious. Scientists at Switzerland's EPFL began with a rat neocortical column—a tiny cylindrical section containing about 10,000 neurons. Using supercomputers, they mapped every cell, every connection, every molecular interaction. The idea: understand the building blocks, then scale up to mice, then primates, then humans.

This wasn't just theoretical modeling. Researchers collected massive amounts of real experimental data—the shapes of neurons, the behavior of ion channels, the patterns of electrical activity—and used it to build biologically accurate digital replicas that behaved like living brain tissue.

Reality Check: 86 Billion Problems

The human brain contains 86 billion neurons, each with an average of 7,000 connections to other neurons. That's roughly 600 trillion synapses, all constantly changing in response to experience. Simulating this at the cellular level would require computational power measured in 4 × 10²⁹ TFLOPS (floating-point operations per second).

For context, Japan's Fugaku supercomputer—one of the world's most powerful—peaks at about 5 × 10⁵ TFLOPS. We're not in the same ballpark. We're not even in the same sport.

Current predictions, based on trends in computing power and neuroscience techniques, suggest:

• Mouse whole-brain simulation: around 2034
• Marmoset (small primate) brain: around 2044
• Human brain: well beyond 2044, possibly decades later

What We Actually Learned

Despite missing the 2023 deadline, these projects achieved remarkable breakthroughs. In 2023, researchers used the Human Brain Project's infrastructure to simulate different states of consciousness—modeling the transition between deep sleep and wakefulness at multiple scales simultaneously, from individual synaptic receptors to whole-brain activity patterns.

This work has practical applications for understanding anesthesia, coma patients, and neurological disorders. Scientists can now predict how microscopic changes at the synaptic level ripple outward to affect consciousness itself—experiments that would be impossible or unethical in living subjects.

The projects also proved something important: partial brain simulations work. They accurately reproduce real brain behavior in specific regions, validating the reverse-engineering approach even if the full-scale version remains distant.

The Consciousness Question

Would a complete cellular brain simulation be conscious? That's the million-dollar question—or rather, the billion-euro question. The Human Brain Project generated heated debate about whether simulation equals experience. Some philosophers argue consciousness requires biological substrate; others suggest it's purely about information processing patterns, regardless of hardware.

We might find out the answer. Just not by 2023. The quest to build a brain continues, one simulated neuron at a time.