Jülich's Supercomputers Target 20 Billion Neurons; Skeptics Snub 'Fundamentally Impossible' Brain Simulations
The Jülich Research Centre aims to model the human cerebral cortex complexity, preparing to simulate 20 billion neurons and 100 trillion connections using the JUPITER supercomputer.
Advocates claim simulations let researchers test impossible-to-study theories, like modeling epilepsy. Critics, however, push back hard. Professor Thomas Nowotny warns that reaching correct simulation scale does not mean one can actually build brains. The conversation pivots between sheer computational scaling and fundamental biological limits.
The overwhelming thrust pushes massive computational power toward replicating brain structure, built on successes like the 2024 fruit fly circuit map. The core schism remains: whether computational scale is the ultimate limiter or if physical biological constraints are insurmountable, regardless of hardware.
Key Points
The capacity to simulate 20 billion neurons and 100 trillion connections using JUPITER.
The Jülich team asserts this massive scale is technologically achievable for modeling the human cortex.
Simulation only tests theories impossible with real brains.
Markus Diesmann notes that this allows study of function otherwise inaccessible, citing disease modeling.
Simulating complexity does not equal building reality.
Professor Thomas Nowotny states it is 'fundamentally impossible to build brains,' regardless of simulation accuracy.
The advancement builds directly on established biological mapping.
The recent completion of the fruit fly brain circuit map (2024) sets a measurable precedent for this ambition.
Source Discussions (3)
This report was synthesized from the following Lemmy discussions, ranked by community score.