GM Readers! ☀️

Welcome to the 9th issue of Evolving Internet Insights — a weekly newsletter curating the top stories and our insights on the “Evolving Internet,” which covers AI, Web3 and everything in between.

Quote we’ve been saying a lot this week: “Every abundance creates a new scarcity” (Chris Anderson, Head of TED, the non-profit behind TED Talks)

Very fitting quote to describe the AI revolution.

Thanks for reading!

Liang and Dan 🙌

🌰 In a Nutshell

• The ARM IPO might be the biggest US IPO since 2021

• X/Twitter confirms it will train its AI models on user data.

• Walmart launches its internal generative AI assistant to super charge employees.

• ARK predicts that eventually everyone can train their own AI model at low costs.

💾 Byte-sized Stories

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

1. All Eyes Are On The ARM IPO

⚡️ TL;DR: A group of investors that include the largest AI companies like Nividia, Apple, Google are looking to invest in the upcoming ARM IPO. ARM designs chips used in 99% of the world’s smartphones. (It doesn’t manufacture chips themselves). ARM is aiming to go public at a $52B valuation, which would make it the largest US IPO since 2021.

⚡️ So What: It is one of the most anticipated IPOs of the year and could end up becoming a litmus test for some of the market’s biggest questions:

• Is the IPO market back? The IPO markets have been slow lately, a successful IPO could signal that there is retail appetite for more IPOs. Companies like Instacart, Klayviao and Stripe are all in line to go public after ARM.

• Is Softbank back? After a rough year of losses, Softbank is hoping that the ARM IPO rides the AI wave and generates healthy returns for its investment. Softbank paid $32B for ARM in 2016.

• Is the AI hype real? Perhaps the most important question on everyone’s mind is whether the AI boom has real staying power. A successful ARM IPO would go a long way in signaling that AI is more than hype.

In short, there are a lot of eyes on ARM. 👀

Read More Here, Here

2. X Confirms They Will Use X Data To Train Its AI Models

⚡️ TL;DR: X (f.k.a. Twitter) confirmed in its privacy policy that it will use the public data on the X platform to train its AI models. The specific section from the privacy policy reads, “we may use the information we collect and publicly available information to help train our machine learning or artificial intelligence models for the purposes outlined in this policy.” This follows a larger announcement by X to collect biometric data from users as part of its attempt to rid the platform of bots and fake accounts.

⚡️ So What: As we’ve covered in a previous issue, Musk launched an AI company called xAi. He has publicly stated that he wants to train xAi models on top of X data. By some estimates, X generates 500M X/Tweets per day, which is an insane amount of data that Musk has all to himself and his companies. Interestingly enough, Musk recently tweeted a call-to-action for more journalists to publish on the X platform (read: generate more data).

⚡️ Zoom Out: X has shown its staying power even in light of competing platforms like Meta’s Threads app – the latter had a meteoric rise in users but retention and engagement have been down since launch. This highlights how X data represents a valuable source of proprietary competitive advantage in the AI race… despite all the complaints, users are still using the platform.

Read More Here, Here

3. Walmart Launches Its Generative AI Assistant

⚡️ TL;DR: Walmart, the largest retailer in the world, rolled out its generative AI assistant to 50,000 corporate employees. The tool, called My Assistant will help employees automate tedious and repetitive tasks like summarizing documents and draft writing. It can also be used by employees to answer questions about Walmart, like helping new hires better understand their company provided benefits.

⚡️ So What: Walmart has been experimenting with and implementing AI solutions to augment its workforce for some time. For context: Walmart is one of the largest employers, employing over 2M associates worldwide, in addition to a corporate workforce of 50,000. The company has integrated conversational AI to help customer service teams better serve customers. In-store, employees can use the “Ask Sam” voice assistant to help them locate items and checkout counters use AI powered cameras to minimize shoplifting.

⚡️ Zoom Out: We are starting to see more and more companies launch internal generative AI tools (built on top of their data) to help employees upskill. Last time we covered how consulting giants like McKinsey and PwC are using AI. In the Age of AI, an upskilled workforce that knows how to use AI is a competitive moat.

Read More Here, Here

📊 Let’s Get Graphic

One visual we couldn’t stop thinking about.

Ark Invest Predicts a 99.99% Decrease in AI Training Costs Between 2020-2030  🤯

⚡️ Takeaway: Costs to train a model will continue to decrease, ARK estimates that by 2030, the cost to train a model as good as GPT3 will drop to $30, a 99.99% decrease during the period. And when you consider that computation is around 100 times cheaper [today] than it was in 2000, the rate of declining costs really frames the potential and accessibility of this technology over the long term.

This means that almost anyone can train their own models which could lead to a commoditization of models. This further makes the case that proprietary data will be critical to differentiate.

🧠 Brain Food

One focus topic to feed your brain.

In the next 🧠 Brain Food newsletter hitting your inbox on Tuesday, we will cover how CoreWeave, an AI cloud computing provider, is competing with existing industry giants like Amazon, Google and Microsoft 👀

Subscribe to get it straight to your inbox on Tuesday, you’ll want to read this David versus Goliath story ⚔️🛡

🐇 Down the Rabbit Hole

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

1. Why AI Is The Mother Of All Unicorns: Blog musing on the winners and losers in AI and how the market might be segmented in the future

2. Why Spotify Struggles to Make Money: Great deep dive into Spotify’s business model and why it struggles to make money

3. Generative AI and Intellectual Property: Essay outlining the input versus output relationship for generative AI models with philosophical questions around who owns what in the Age of AI.

🌊 Watercooler Talk

Overheard from our community.

In some of our circles this week, we heard chatter about what happens when the world is mostly AI generated content? Experts predict that 90% of online content will be AI generated by 2026.

Here were some takes:

• AI Models will increasingly be based on AI generated content, what happens then?

• As we spend more time online, how will we discern what is real?

• The concept of “real” will evolve, but maybe this is a logical evolution because most of our lives are already curated by algorithms anyways.

P.S. Got some watercooler talk for us? Share your suggestions in our comments section!

📝 Elite Jobs Corner

From our networks, we curate and connect top founders and companies to rockstar recruits.

⚡️[Web3] Founding Engineer at Karma3 Labs

• What: Karma3 Labs is building the reputation layer for the Internet. As more of our lives continue to move online, knowing who can be trusted and who cannot will be critical.

• Who: Founding team has experience launching and scaling Web2 and crypto startups with successful exits. Deep technical and research talent. One of the Cofounders wrote the algorithm that powers how Google ranks pages. 🤯

⚡️[AI] Founding Lead Infrastructure Engineer at Konko.ai 

• What: Konko AI is a fully managed API that allows developers to easily access, evaluate, choose, fine-tune and deploy large language models (LLMs)

• Who: Founding team with operator and technical experience in AI, venture backed startups, consulting, both HBS grads 🤓

📣 Call to Action

Are you a potential candidate? You can apply directly and reply to this newsletter with an email asking for us to refer you (please include why we should refer you). What are you waiting for? Take a chance. 🎲

Are you hiring? If there is a rockstar recruit that you want to get connected to and/or you want us to showcase the role(s) you are hiring for, please reply directly to this newsletter. We would love to hear from you! 👂

Do you know someone who is a great fit for a role? Reply to this newsletter providing us more details in your email (also, since you are already in your inbox, forward them this newsletter and ask them to subscribe). 🙏

🗳 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 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).