Ushering in the Thermodynamic Future - Litepaper
We are very excited to finally share a bit more about what we are building: a full-stack hardware platform to harness the natural fluctuations of matter as a computational resource for Generative AI.
The demand for computing power in the AI era is increasing at an unprecedented exponential rate. Luckily, for the past several decades, the miniaturization of CMOS transistor technology following Moore’s law [1] has allowed much of this exponential growth to be accounted for by increasing computer efficiency. Unfortunately, Moore’s law is starting to slow down [2]. The reason for this is rooted in fundamental physics: transistors are approaching the atomic scale where effects like thermal noise start to forbid rigid digital operation [3][4][5]. As a result, the energy requirements of…
saved by
- Yufei Xiao
- Svitlana Midianko
- Tazik Sh
- Rajan Agarwal
- Turja Chowdhury
- Hangyul Lyna Kim
- Yoyo Yuan
- Aishwarya Mahesh
- Mackay G
related reading
- Thermodynamic Computingarxiv.org
- Vibe physics: The AI grad student \ Anthropicanthropic.com
- An efficient probabilistic hardware architecture for diffusion-like modelsarxiv.org
- Computational Foundations for the Second Law of Thermodynamics-Stephen Wolfram Writingswritings.stephenwolfram.com
- Introducing Un-0: Generating Images with Coupled Oscillatorsunconv.ai
- On the Biology of a Large Language Modeltransformer-circuits.pub
- Tiny TPUtinytpu.com
- Join Us in Building the AI Stack for Industrial Engineering at Mistrallinkedin.com
- Fluid dynamics feels natural once you start with quantum mechanicsyoutube.com
- Exploring a space-based, scalable AI infrastructure system designresearch.google
- Hemispherical Stacks | Hemispherical Stackshemstacks.antikythera.org
- AI Chip Architecturesjacobpeake.com