X's new Product Lead Peter Sellis explains why he designs teams "like a terrorist organization": "I should probably workshop this one a little bit. I think it comes from essentially two primary beliefs. One is that there's actually a hidden cost to collaborating that no one really internalizes. You'll never really hear any executive at any company be like, stop collaborating. And to me that's kind of like a warning sign." "Collaboration comes with incredible coordination costs. And coordination costs essentially assure that you're moving as slow as the slowest node in the system." "The second thing is that at high growth companies especially, product managers should be in a position where they're making decisions that the founders would have made, sometimes 6 months ago, definitely a year or two ago." "Set aside violence for a moment, I'm not saying that I designed my teams for violence, but a good terrorist organization essentially has 2 things. One is this ideological culture that everyone understands why they are doing what they're doing. And then it has a very, very clear organizational structure of who can be trusted with which decision." "The combination of those 2 things can be leveraged in a way such that you don't need to over-collaborate, and you can be assured of achieving the results that you wanted as if you had collaborated."

Dario on finding meaning when AI is the best at everything: chess got more popular after Deep Blue beat Kasparov, not less. “Meaning is a really subtle thing. If AI is the best at everything, does meaning only come from being the best in the world at something? I think that’s the question being asked here.” “Something I find interesting is that long after Deep Blue beat Kasparov, there’s been a renaissance in interest in human chess players. Magnus Carlsen is like this phenomenon. I think he’s like a chess player and a fashion model, but he’s like a meme.” “If AI threatened meaning, you would think that all of this would go away 30 years ago, after Deep Blue beat Kasparov. But if anything, it seems to have stoked even greater interest in humans playing chess. So I think this is some kind of parable for the post-AGI age.”

Dario says Anthropic stress-tests its models like a CIA recruit to catch lying, deception and sycophancy "On the technical side, we've gotten very good at exposing the model to these extreme stress tests, where we put it in these very difficult situations to try and ferret out lying or deception or sycophancy." "You can think of it as almost analogous to the test I imagine you have to go through if you're trying to join the CIA. You just put the model in these difficult situations and try and rat out if it's disloyal or going to do the wrong thing. It's a weird area of technical innovation, but it's one in which we've excelled."

Dario on why autonomous weapons worry him, and why Anthropic drew red lines with the Pentagon: "On autonomous weapons in particular, we had two more specific concerns. The first concern was simply that AI can be unpredictable and unreliable. It's still a developing technology. So one point is simply the technology isn't ready and isn't reliable enough." "There's a second concern we have, which is what you might call oversight or accountability. Suppose these things were fully reliable. So then you manufacture an army of a million robots or a million drones. It's a little unlike human soldiers, isn't it?" "With human soldiers, they're supposed to follow this thing called the Constitution, and if they get orders that contravene the Constitution or that just seem morally abhorrent, then they can just refuse to follow orders. Whereas the AI models, that isn't true." "The spectre of one person having their finger on a button that controls a million robots or a million drones, it's kind of fundamentally undermining to democracy unless you have the right guardrails on it. And our law hasn't really caught up."

Chamath says a generation raised on TikTok instead of trades is why America can’t build AI data centers fast enough Dave: “Elon was saying there’s 15 gigawatts of chips just waiting for liquid cooled data centers to fall into. That’s a massive asset depreciating idly.” Chamath: “It’s worse than that. There’s also the non Vera Rubin stack chips, the older generation that are also still sitting. There’s a bunch of these guys that have these things sitting in inventory. Those aren’t being used.” “Everybody thinks this stuff is easy and it turns out, back to Elon, as he’s proven, physics is really hard and mechanically building things is really hard.” “It’s probably particularly hard because we’ve spent 20 years teaching the younger generation of kids who should have been learning how to do these trades just to be on TikTok and social media and video games. And so the ability to actually build these things is gone.”

Michael Saylor makes the case for an 8%-yield Bitcoin-backed stablecoin over the $350B of stablecoins paying nothing "Who doesn't want a stablecoin that pays 8% instead of zero, or wouldn't you rather have a money market that pays you 10 instead of 4?" "After talking to all of our partners, people that are doing it like Apex and Saturn, it's pretty clear it's beneficial for them if we go to daily because we strip that vol." "The next step is how do you create digital money? And digital money would be pegged to the dollar, 0 vol, paying 6, 7 or 8%. That's the mythical Bitcoin-backed stablecoin that pays you 8%." "And if you had a stablecoin that paid you 8%, why in the world would you hold $350 billion of stablecoin that pays you nothing?" "We're all chasing after that money market instrument where you could buy $100 million of it on a Thursday, collect 8% and then get liquid and get all your money back on Monday." "That makes it perfected money. It's a medium of exchange, unit of account, store of value, and you can use it for payments, collateral, trading, and you can hold it."

Threadguy says appetite for pure memes is at an all-time low, but trading experiments like papertrade and Cards are seeing huge demand "Maybe I'm hating because I missed it. Shout out to Rasmr for banking the Kimchi trade. Shout out to Malcolm for banking it as well." "But it does in fact feel to me like a significant lower high on an Ansem coin type of thing. To be fair, it's a little bit different. He didn't endorse it and tweet about it and do the whole thing." "But willingness to participate in purely memes right now is ATL. ATL. ATL. All-time low." "On trade experimentation stuff, I think there's huge appetite. You look at papertrade, you look at what Jack Butcher did, you look at what Cards is doing. There is huge appetite for that." "But purely memes, people really don't want to play right now. They really don't want to play."

Cameron Winklevoss on his $ZEC thesis: if privacy is worth 10% of Bitcoin, it's a 5x from here "One of the easy ways to sort of think about Zcash is also in reference to Bitcoin. Right now, I think the market cap of Zcash is somewhere around $25 billion and I think it's about 2% of the size of the Bitcoin market cap." "So, what is the value of privacy? Is it 10% of Bitcoin? We love Bitcoin, and I don't think it's going anywhere, but I think that privacy is probably worth more than 2% of Bitcoin." "Let's say it's worth 10%, then you have a 5x. You can sort of back into that very easily, and you find yourself easily above $5,000 a coin. And there's clearly a path to, I think, a $10,000 coin."