Gm. Happy Monday everyone. Contrary to popular opinion, I’ve always been a fan of Mondays. But if you’re a normal Monday-hater, I hope this newsletter makes the day a tiny bit more tolerable.

— Stephen Flanders

Besides being an amazing piece of technology, GPT-4 was also a heavily hyped piece of technology. Who could forget the infamous GPT-4 100T memes?

Eventually, the hype got so intense that Sam Altman had to warn people they are “begging to be disappointed”.

Thanks to some new leaks, we now know what he was talking about.

If George Hotz (founder of self-driving startup Comma.ai), Soumith Chintala (co-founder of PyTorch at Meta), and Mikhail Parakhin (Microsoft Bing AI lead) are to be believed, then we don’t know what we thought we knew about GPT-4.

Previously, it was believed that GPT-4 was one big model. Instead, according to these esteemed gentlemen, GPT-4 is not one big model but eight smaller (~220B) put together.

Buy why does this matter? GPT-4 is really damn good, who cares if it’s not one big model?

For this explanation, we turn to our favorite AI blog: The Algorithmic Bridge (TAB).

OpenAI has ascended into almost mythical status due to the success of ChatGPT. As you can imagine, the mythical among us are afforded many perks, including the opportunity to shape regulation.

OpenAI needed GPT-4 to appear magical, even if it actually wasn’t the breakthrough that people were anticipating. So they shrouded it in secrecy and let people’s imaginations go wild trying to figure out what was behind the curtain.

As TAB puts it, GPT-4 is magic, just the kind of magic we see at a show, not the kind you’d find at Hogwarts.

The Bottom Line: GPT-4 is very good. But it might not be as revolutionary as previously believed.

The spread of AI has even more people than usual wondering what the future will look like. Will we live in an AI utopia, as Marc Andreessen believes? Or will we die a painful robot-assisted death, like Eliezer Yudkowsky believes? Or (most likely), will the reality be somewhere in the middle?

Well, Pew Research Center took the time to interview the experts to see their opinions on what the world will look like in 2035. It’s a super interesting read, but it’s also quite long, so here are the main points:

There’s a lot to take in, but if you have time, we really recommend reading the entire report.

The Big Picture: Nobody knows what’s going to happen tomorrow, let alone in 2035. However, it’s still helpful to ponder the future so that we can build the best technology possible.

What else is going on?

Stat: $100M: The amount of money Amazon is putting into its Generative AI Innovation Center. The innovation is going to be absolutely flowing.

Rabbit hole: AI and The Burden Of Knowledge (The Generalist)

Just get stuff done.

How much is FTX seeking to claw back from SBF’s shills “super networkers”?

The best resources we came across this weekend that will help you become a better founder, builder, or investor.

🕹️ Ctrl saves you hours by doing your customer-related busywork for you.

📈 The Virtual CMO solves any marketing problem in less than a minute.

🚨 If you’re a solopreneur, avoid these things at all costs.

Mark “The Terminator” Zuckerberg

If you’re interested in advertising with us, send an email over to [email protected] with the subject “Homescreen Ads”.

$700M. That’s a whole lot of money that wasn’t being used very altruistically.