Will AI save us or destroy us?

For Eliezer Yudkowsky, the day OpenAI launched, the world ended. “That was the day I realized that humanity probably wasn’t going to survive this,” he said on a recent podcast with Ezra Klein.

For the uninitiated, Yudkowsky is no fringe voice. He founded the Machine Intelligence Research Institute. He helped define the “alignment problem” — how to make sure superintelligent systems share human values. Wherever you land on the AI spectrum — doomer, accelerationist, or just uneasily curious — his words, I think, are worth reading.

Yes, technology has lifted billions and reshaped civilization. But humans have a terrible record when it comes to hubris. I don’t share his certainty that we’re doomed. But I do share his suspicion that history is, once again, repeating — brilliant people, good intentions, and total blindness to consequence.

Key quote: “Back in the day, people would always say: Can’t we keep superintelligence under control? Because we’ll put it inside a box that’s not connected to the internet and we won’t let it affect the real world at all, unless we’re very sure it’s nice… But in real life, what everybody does is immediately connect AI to the internet. They train it on the internet. Before it’s even been tested to see how powerful it is, it is already connected to the internet, being trained. Similarly, when it comes to making AIs that are easygoing, the easygoing AIs are less profitable. They can do fewer things. So all AI companies are throwing it harder and harder problems because those are more and more profitable, and they’re building the AI to go hard on solving everything, because that’s the easiest way to do stuff.”

When thinking slower is the actual edge

Clearly AI has been on my mind lately: I was recently interviewed by John Rotonti of Bastion Fiduciary for his podcast Rebellious Allocations, where we talked about AI, field research, and how I’m using — and deliberately not using — AI in my own investment and research process.

For me, the challenge isn’t how to be more productive or move faster using AI. It’s the opposite: learning to slow down, and to think, observe, and ask better questions in a world built for acceleration.

You can listen to the full conversation here.

Key quote: “You can do an AI summary of a transcript; you can do an AI summary of a 10-K. I think the irreplaceable value comes from in-person, real conversations. We’ve integrated AI into our process, but we’ve also prioritized more in-person research trips because that’s where you increasingly find an edge. Go spend time at the company, interview people around it, use the products, invest yourself in the ecosystem. You can’t replicate that by just listening to an earnings call.”

A few more links I enjoyed:

Barry Diller – Instinct Over Algorithms – via The Knowledge Project

Key quote: “‍I don’t like it when people come in to the early stages of a broad, general entertainment career and say, ‘I want to run a studio’ or things like that. I think whatever your interest is, you just get on the broad path. Whatever door you’ve got to bang to get in there, you get on the broad path, and everything else takes care of itself. You don’t have to say, ‘I want to be this’ or ‘I want to be that,’ and I’ve always mistrusted that.”

Why 2025 is the single most pivotal year in our lifetime – via Big Think

Key quote: “‍We are living through the collapse of the old world, and the quiet construction of a new one. From artificial intelligence and clean energy to bioengineering and digital governance, the core systems that defined the last century are rapidly being dismantled and replaced. But this isn’t just about technology. According to futurist Peter Leyden, we’re at a historic turning point: One of the rare moments in American and global history when everything gets reimagined at once.”