The Lightcone 2025 Forecast

Welcome back to a New Year's Day episode of the Light Cone. We wanted to hit you guys with predictions on what's gonna happen in 2025. Diana, do you wanna start us off? Sure. So last year, AI won two Nobel Prizes. The first one was the Nobel Prize in Physics with Jeff Hinton on his seminal work on deep neural nets.

And the second one was a Nobel Prize on chemistry from Dennis and his team on alpha fold and protein folding. And we're seeing how AI is really accelerating abundance with discoveries with science and technology. And there's a total of six Nobel prizes. There's four more to go. So why not winning one more? It could be the one on math perhaps. It could be interesting.

I know there's been a lot of collaboration with OpenAI and Terence Tao, the famous mathematician, and a bunch of researchers trying it too. Could be one in economics. I don't know if we come up with some interesting economics theories, but maybe those. I think the peace one in literature might be a bit more tricky. Alright. My one is a dangerous one to make on camera.

It's that next year, crypto is going to go mainstream. Very specifically, my prediction is this time next year, like, you and everyone you know will have bought at least one thing during the year that's, like, real. Like, whether it's like a cup of coffee or a book using a stablecoin.

And the reason I think this will happen is payment when you think about it, like, payment networks are actually just marketplaces. The marketplace businesses just have, like, vendors on one side and consumers on the other. And we all know from our work with startups that marketplace businesses are incredibly hard to start because you have a chicken and egg problem.

But when you do figure out how to crack the chicken and egg problem, you have incredibly sticky enduring business. Like, actually, the biggest YC companies are almost all marketplaces. Airbnb, DoorDash, Instacart. I mean, Coinbase itself essentially started out as a marketplace, though it has like, an exchange is a marketplace, really.

I think what's happened with crypto is clearly everyone came in initially for the speculation and buying things where the numbers went up. But stablecoins have emerged as, like, a truly, like, valuable use case. And so many, many people have a crypto wallet, which at this point is hundreds of millions of people, will recognize that stablecoins are a really great way to just buy things.

And so you basically have one side of the marketplace lined up. There's hundreds of millions of people who will, like, have stablecoins in their crypto wallets. All you need now are the merchants to start accepting USDC or a stablecoin. And I think that's actually the much easier part of the marketplace to get. Yeah. There's a lot fewer of them. Yep.

There's fewer of them, and it's just a more traditional ground sales game. Yeah. And I think people especially now the regulatory environment is looking more optimistic. I think next year, you'll just see a big push, and you'll walk into your local coffee shop, and they'll accept USDC payments. Think it should be huge.

It's pretty clear that stablecoins are, you know, sort of desperately needed in finance overall. I mean, when you look at something like that Synapse disaster, that startup that lost, you know, tens to hundreds of millions of dollars of user funds, know, if you had, you know, Open Ledger, I think that you would be able to figure it out.

But right now, they don't even know, you know, where the money went, and they don't know how to track it down for people. So it's a pretty wild situation when you really think about it. It's overdue.

So I think for mine, I also have one that is kind of crypto related in that I think if Doge, the, you know, sort of Elon Musk and Vivek attempt to reduce government waste works, then Dogecoin will go up in a very circuitous sort of way.

You know, I think that interest rates are largely sort of driven, you know, by a lot of different factors, but one of the bigger ones is actually whether or not the government is actually doing a lot of deficit spending.

And I think the markets, the way they're moving, at least right now, as we're, you know, watching it in late December, you know, they think that Trump is probably gonna spend a lot more money. But, you know, if Doge is able to actually reduce spending, well, that means that interest rates will come down and, you know, sort of a, hopefully, the economy will actually pick up.

There will be more g GDP growth, but also interest rates rule everything around us as we've learned the last, you know, three or four years. So, you know, sort of concretely, crypt as crypto prices go, they they turn out to be highly correlated to interest rates. So interest rates go down, crypto prices, and hopefully GDP goes up.

And, you know, likewise, it's probably true that the reverse is also going to happen. If Doge fails in their mission, it seems pretty clear that Trump left to his own devices would probably increase, you know, spending the way he did in his first term. And so I don't know. I mean, what can I say? We're we're rooting for you, Elon. Yeah.

I think another aspect that ties into what we've been talking about is the worry with lower entry lowering interest rates is that inflation would spur back up. But tech in general is deflationary, and AI in particular may be, like, the greatest single greatest deflationary force of our lifetimes.

And so if that plays out, you may well be able to lower interest rates without stimulating huge amounts of inflation. Alright. I've got an AI one, which is a product that I'm just, like, personally excited to see and play with that I think is gonna come out next year. And it's basically you will be able to have a Zoom call with an AI.

Like, in 2024, that was the year where you could finally have a phone call with an AI, and they got the latency low enough that it felt, like, natural and real.

And I think next year, we're gonna be able to do the same thing, but it's gonna be like a Zoom call where there's, like, a face and a and a whole, like, virtual avatar, but it's actually, like, real time, and you're interacting with, like, a a virtual, like, AI person on a screen. That doesn't feel Uncanny Valley. Because right now, there are different versions of it that are very unkind.

Latency is bad, and the lip syncing is bad, and it's just like it's just like not a good experience yet. But I think it's gonna yeah. A three d Turing test. Gonna happen. Yeah. Like a three d Turing test. Yeah. Maybe we'll get them to do the Litecoin episodes for us.

Yeah. Surprise. This is actually all AI. Just kidding. It's still real. It's still real. So with that, we just wanted to give you a lightning episode of our predictions for 2025. Happy New Year, and we'll catch you guys next time.