YouTube · Google · IEEE Spectrum AI
Today, you probably asked a question of a large language model, or accepted a connection suggestion on LinkedIn
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
◌ Single Source
But what you might not know is how much energy that interaction consumed or why.
Key facts
- The MOSFETs were created in a commercial foundry using fabrication technology called the 180-nanometer node, which was cutting-edge in the year 2000
- Working in their laboratory in 2024, one of their students was measuring a memory circuit that consisted of one transistor and one memristor—a type of nonvolatile memory device first fabricated in 2008
- But each of those GPUs achieves that by consuming as much as 1,000 watts apiece
- Let’s say this “threshold voltage” is 0.7 volts, although the real number depends on device geometry and silicon doping
Summary
The result is a simple and efficient neuromorphic device that mimics a brain cell. Today, you probably asked a question of a large language model, or accepted a connection suggestion on LinkedIn, or watched a recommended video on YouTube, or took a different route to work based on a traffic prediction from Google Maps. AI requires processing massive amounts of data, which is usually done in large data centers populated by thousands of GPUs capable of executing up to trillions of operations per second. Fundamentally, a lot of this inefficiency is because GPUs are trying to simulate the workings of artificial neural networks using software and billions of transistors, which requires using energy to move massive amounts of data. The brain is roughly one million times as energy efficient at many of the comparable tasks they set for AI.