How Reinventing the Transistor Could Revolutionize AI Hardware

In this video, created by Scixel (https://scixel.es/), you can discover the remarkable journey of computing—from the birth of electrical circuits and Boolean logic to the invention of the MOSFET transistor that made modern computers possible.
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Today, artificial intelligence depends on billions of transistors, but this incredible computing power comes at an enormous energy cost. Could there be a better way? In this video, we explore a groundbreaking 2025 discovery that transforms a single silicon transistor into an electronic neuron by harnessing impact ionization. 

This innovation could pave the way for low-power neuromorphic computing, bringing powerful AI to smartphones, robots, autonomous devices, and the Internet of Things. Topics covered: 

The history of the transistor 
Boolean logic and the birth of computing 
Why AI consumes so much energy 
Neuromorphic computing explained 
The single-transistor silicon neuron 
The future of low-power artificial intelligence 

If this technology succeeds, it could become one of the most important advances in computing since the invention of the MOSFET. 

More information about the single-transistor neurons: 
https://www.nature.com/articles/s4158... 

More information about the developing team: 
https://lanzalab.org/research 
https://www.newmorphic.com/

#ArtificialIntelligence #Transistor #MOSFET #NeuromorphicComputing #Electronics #Semiconductors #Technology #Engineering #ComputerScience #AI

Follow the Topic

Artificial Intelligence
Mathematics and Computing > Computer Science > Artificial Intelligence
Electronics Design and Verification
Technology and Engineering > Electrical and Electronic Engineering > Electronic Circuits and Systems > Electronics Design and Verification
Electronic Devices
Physical Sciences > Physics and Astronomy > Condensed Matter Physics > Electronic Devices
Electronic Devices
Physical Sciences > Materials Science > Materials for Devices > Electronic Devices
  • Nature Nature

    A weekly international journal publishing the finest peer-reviewed research in all fields of science and technology on the basis of its originality, importance, interdisciplinary interest, timeliness, accessibility, elegance and surprising conclusions.

Related Collections

With Collections, you can get published faster and increase your visibility.

Carbon Dioxide Removal

In this cross-journal collection, we showcase articles that help with understanding how carbon dioxide removal can contribute to climate change mitigation.

Publishing Model: Hybrid

Deadline: Jan 16, 2027

Cancer Neuroscience: from mechanisms to therapy

With this collection, Nature Communications, Nature, Nature Cancer, Nature Neuroscience, Nature Progress Oncology, Nature Progress Brain Health, and Communications Biology invite submissions that aim to elucidate the molecular mechanisms underlying bidirectional cancer–nervous system interactions and their influence on the tumour microenvironment, as well as pre-clinical and clinical studies exploring potential therapeutic strategies to target neural–cancer interactions.

Publishing Model: Hybrid

Deadline: Jan 30, 2027