Sama Bored with Building AGI, Wants to be Paul Graham

This is The Startup Breakdown, the newsletter where we learn, laugh, and love startups. By joining this growing community of hundreds of future startup aficionados (think i spelled that right?), you're getting a beachside view of the ocean that is the startup and VC scene. This ain’t your grandpa’s newsletter, so prepare yourself for an inbox full of 4/20 jokes and Succession references.

Happy Wednesday, folks.

In our Infobot Weekly shows, we have started to ask guests 2 questions:

• What is one piece of news from your area of expertise that nobody outside of that field is talking about?

• What is your most controversial opinion about your field?

Not going to lie, some of the answers have been crazy fascinating. If you have a good answer to either/both of these, either comment at the bottom of the email or reply to this email directly 👀 

Sama Bored with Building AGI, Wants to be Paul Graham

Even robot knees can get worn down, apparently, so ChatGPT is hanging up the cleats and picking up the whistle to join the coaching ranks, recruiting and developing the next batch of All Star AI startups.

Enter Converge 2, OpenAI's six-week boot camp that's part startup accelerator, part Hogwarts for former Rocket League players.

OpenAI's vision is to identify and cultivate the next generation of AI-driven startups, redefining a variety of industries with the OpenAI team’s tutelage and a golden ticket worth $1 million.

Through a pool of tech talks, social events, and Blue Bottle chats with AI bigwigs, likely looking to create a YC-like experience (and Sam is obviously very familiar with the org) for AI founders.

Opening an in-house venture arm is a rite of passage for companies making the transition into womanhood, much like girls realizing that Starbucks Frappuccinos are objectively terrible and not even real coffee. In embedding themselves into the very DNA of the next AI revolution, the company is further embracing its pick and shovel-like status in the space.

However, though you’d think that the know-how and resources of these tech giants would lend themselves to high-upside companies, the consensus is that they’ve traditionally been less successful than traditional VCs, possibly an indictment on the companies’ ability to identify talent (at least that’s what I keep telling myself re: my FAANG rejection letters), but also possibly the result of misaligned incentives between the investing company and the investee, where the bigger company “nudges” the startup into building products and services which provide strategic benefits to the bigger company rather than optimizing for growth and high returns.

It has been well-established that the pool of top-tier AI talent is fairly limited, thus when one of these few dozen individuals asks for money, there are lines down the hallway to give it to them. OpenAI claims to be taking a different approach, saying that founders don’t have to have some fancy pedigree or come from a 49ers family. To be honest, until there is ample evidence to say otherwise, I really don’t anticipate the batch’s founders to look any different than the MIT/Stanford CS grads that make up every other successful startup incubator.

OpenAI also claims that there is no need for participants to build on OpenAI’s APIs, and while I am 99% certain that this is true and they’re not just explicitly lying, I can also be pretty confident in conjecturing that the company will make the perks of doing so akin to holding a gun to their head.

So, is OpenAI the new Don Corleone of the tech world? Maybe.

But what is certain is that this marks a significant step in OpenAI’s growth evolution, and if Sama can put aside his own personal ambitions for OpenAI (unlikely) and adopt the same mindset that helped him to lead Y Combinator, we could see a new player emerge in the investment game, hopefully creating even more world-changing AI startups.

Cheers to another day,

Trey