From CNBC.

Nvidia’s blowout earnings Wednesday affirmed its place as the world’s most valuable company, thanks to “off the charts” sales of GPUs that excel at AI workloads. But as inference needs outpace training, a smaller cheaper AI chip is gaining ground. Custom ASICs are now being designed by all the major hyperscalers. Google led the way with its TPU, followed by Amazon, Meta, Microsoft, and OpenAI’s plans with Broadcom. CNBC talked to experts and insiders at the Big Tech companies making them all.

Chapters:
0:00 Introduction
1:31 GPUs for general compute
5:07 ASICs for custom cloud AI
10:29 On-device chips for edge AI
13:21 Geopolitics and power

Produced and Shot by: Katie Tarasov
Edited by: Evan Lee Miller
Additional Camera: Marc Ganley, Jordan Smith
Senior Director of Video: Jeniece Pettitt
Animation: Jason Reginato
Editorial Support: Kif Leswing
Additional Footage: Amazon, AMD, Apple, Broadcom, Cerebras, Getty Images, Google, Groq, Ford, Intel, Meta, Nvidia, Qualcomm, Samsung, TSMC

» Subscribe to CNBC TV: https://cnb.cx/SubscribeCNBCtelevision
» Watch CNBC on the go with CNBC+: https://www.cnbc.com/WatchCNBCPlus

About CNBC: From ‘Wall Street’ to ‘Main Street’ to award winning original documentaries and Reality TV series, CNBC has you covered. Experience special sneak peeks of your favorite shows, exclusive video and more.

Want to level up your AI skills? Sign up for Smarter by CNBC Make It’s new online course, How To Use AI To Communicate Better At Work. Get specific prompts to optimize emails, memos and presentations for tone, context and audience. Sign up today: https://cnb.cx/47ELu8O

Connect with CNBC News Online
Get the latest news: https://www.cnbc.com/
Follow CNBC on LinkedIn: https://cnb.cx/LinkedInCNBC
Follow CNBC News on Instagram: https://cnb.cx/InstagramCNBC
Follow CNBC News on Facebook: https://cnb.cx/LikeCNBC
Follow CNBC on Threads: https://cnb.cx/threads
Follow CNBC News on X: https://cnb.cx/FollowCNBC
Follow CNBC on WhatsApp: https://cnb.cx/WhatsAppCNBC

#CNBC

Comparing The Top AI Chips: Nvidia GPUs, Google TPUs, AWS Trainium