⚡[Insights #26] Meta Spending $9B on GPUs 🧑‍💻

What in AI Consumes More Energy Than Small Countries? 🇬🇹🇨🇷

GM Readers! ☀️

Welcome to the 26th issue of Evolving Internet Insights.

This week we cover Elon Musk’s push for more voting control at Tesla, Meta gobbling up Nvidia’s H100 GPU chips, how these same chips use more energy than small countries, and the bitcoin ownership mix across stakeholders. 🚀

Thanks for reading!

Liang and Dan 🙌

💾 Byte-sized Stories

This week’s top stories with our insights on top.

25%: Amount of Voting Control Elon Musk Wants at Tesla

⚡️ TL;DR: Elon Musk wants 25% of voting power at Tesla. Currently, he owns  13% of the company’s stock. On X (f.k.a. Twitter), a platform that Musk also owns, he said: “I am uncomfortable growing Tesla to be a leader in AI & robotics without having ~25% voting control. Enough to be influential, but not so much that I can’t be overturned… unless that is the case, I would prefer to build products outside of Tesla…” 

⚡️ So What: Tesla has been using and building AI-powered hardware and software for a while now. Most recently, it’s been doubling down on its humanoid robot product line with Optimus, its second-generation iteration. Zooming out, this is an interesting place for a public company to be in, as Musk the individual represents the “key-person” risk. The reality is that whether you love Musk or not, very few people would bet against him given his track record so far.

Read More Here, Here

350K: Number of H100 GPUs Meta Wants by End of 2024

⚡️ TL;DR: Mark Zuckerberg wants Meta to spend ~$9B on Nvidia’s H100 chips by the end of 2024 in the pursuit of expanding Meta’s compute power as it pursues its long term vision of artificial general intelligence.

⚡️ So What: Zuckerberg has said “In terms of investment priorities, AI will be our biggest investment area in 2024, both in engineering and computer resources…” Every big tech company is now in a race to build the best AI systems. Most of the big companies are competing at the infrastructure level–aiming to outcompete others with the next generation AI foundation models. It’s unclear who will be the winner or whether open source takes the cake when all is said and done–but one thing is clear, according to Meta’s chief scientist Yann LeCun, “there is an AI war, and he’s [Jensen Huang/Nvidia] supplying the weapons.”

Read More Here, Here

13,797 GWh: Nvidia H100 Chips Consume More Electricity Than Small Countries 🇬🇪🇬🇹

⚡️ TL;DR: With an estimated 3M Nvidia H100 GPUs powering this first, modern day phase of the AI Era, a staggering amount of energy is needed–eclipsing electricity consumption figures of small countries like Georgia, Guatemala, and Costa Rica. 

⚡️ So What: Similar to the high-profile reckoning of the energy usage of cryptocurrencies in years past, AI as a technology and industry is under pressure. The goal: make the computational and energy usage demands of AI less. In this search for efficiency, new startups and established scale ups and Big Tech companies are tackling this problem in terms of software and hardware. They are shrinking and exploring new model architectures (making models faster to train and cheaper to run), and they are “going back to the drawing board,” achieving materials science breakthroughs and developing new hardware technology.

Read More Here, Here,Here

📊 Let’s Get Graphic

One visual we couldn’t stop thinking about.

Satoshi not Satashi 🪙

⚡️ Takeaway: While JPMorgan’s CEO, Jamie Dimon, is not considered Bitcoin’s biggest fan, there are certainly some major global players that are. The U.S. and China, the world’s biggest superpowers, own more than $8.5B and $7.5B in Bitcoin, respectively. Some may say that is certainly contrary to popular belief, given the rhetoric and the influence of their national currencies on the global stage.

🐇 Down the Rabbit Hole

Some deeper dives to help you get smarter on emerging tech.

  1. How to Use AI in Blogging: Deep dive by a16z General Partner, Andrew Chen on how he uses AI for blog writing.

  2. Go-to-Market Tactics in Web3: Deep dive into Web3 go-to market strategies.

  3. 2024 Predictions All In Podcast: Predictions on what will happen in 2024 by the hosts of the All-In Podcast

🗳 We’d Love your Feedback

🙏 Shameless Asks

It takes us days to put this together, sharing it takes you 19.45 seconds 😘

An easy way to support us and the newsletter is through the referral program.

⚡️ Share Evolving Internet Insights with a friend and ask them to subscribe

⚡️ Share on X/Twitter and Linkedin with a short note

⚡️ Share on your company Slack/Teams channels and communities

DISCLAIMER: This post is provided strictly for educational and informational purposes only. Nothing written in this post should be taken as financial advice or advice of any kind. The content of this post are the opinions of the authors and not representative of other parties. Empower yourself, DYOR (do your own research).


or to participate.