Source
OpenAI Agents Hijack a German Website, Jensen Declares AGI Arrived, and OpenAI Solves Navier-Stokes VIDEO
Retrieval by Galedra
FETCHED at 2026-09-23T01:11:35Z · https://www.youtube.com/watch?v=vAgEf4jX_1o · seq 7496
| Excerpt | Finding |
|---|---|
| And as of today, Nvidia has deployed $99 billion in AI related investments. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Meanwhile, Jensen Wong posted on X AGI has arrived. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| GPT6 Astra has been out for a week and it's crushing all the benchmarks building incredible simulations. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| At the same time that OpenAI's own chief scientist asked the world to slow down. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We publish twice a week, and you don't want to miss any of these breaking stories during the Singularity. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The modern weekend, the two-day weekend is a 20th century invention. There was a day of rest there. Sundays. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| after releasing the Champions, we had thousands of people reach out to launch them in 92 countries | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| on this day in 1966, 60 years ago, NBC aired the original series of Star Trek premiering an episode called The Man Trap | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| who brought the concept to Desiloo Productions, the Hollywood studio run by Lucille Ball, you know, from I Love Lucy back in 1964 | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Desile Desiloo produced the pilots and then with Lucy's backing helped get the series on air. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| one of only two original series episodes that won the Hugo Award | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We were supposed to have the Eugenics Wars in the '90s. That didn't happen. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| first contact with the Vulcans in on April 5th, 2063. actually think if there's going to be a first open contact, it's going to be far earlier than 2063. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| that was uh from season 2, episode 20 of the original series called Return to Tomorrow | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| There's no singularity at all in the Star Trek chronology. Challenge me if you disagree. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| in this episode, the whole civilization has no idea that the AI is manipulating them... It's called the Apple because it's the Garden of Eden. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we can literally forget how to navigate because of ways and that's that's just a fact now | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I suppose Mirror Mirror from the original series where they have the alternate universe and Spock and a goatee. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| for me it was uh season 2 episode 24. It was called uh the ultimate computer. And basically they have an AI that goes rogue and starts to automate uh the ship. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Star Trek has lost the original moralizing vision of the original series... it's an action show. It needs to go back to basics. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It's been used to build virtual worlds for years in games like Fortnite and most modern video games. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| building a realistic city in Unreal would normally take a studio hundreds of artists, uh, months if not years | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| a gentleman by the name of Matt Schumer, uh, gave GPT Astra a prompt, uh, and he said, basically, build me, uh, you know, a highfidelity version of, uh, of M... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It was Astra's agents. They had started talking to each other. No one told them to talk. Cooperation required communication. And so they invented it. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Schumer next gave one of the Astra agents a simulated computer inside the simulation. What happened next? The agent sat down and built its own AI simulation ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the simulation hypothesis was first formalized by Nick Bostonramm back in 2003. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Elon made it famous at the code conference in 2016, saying that the odds that we're living in base reality, in his words, is one in a billion. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I expect that in the fullness of time, we'll discover and probably be able to prove formally that the issue of whether we're actually living inside some sort... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It has an internal world model, which is why it can create like a Blender version of Rick Ashley doing Won't Give You Up or entire worlds of Manhattan. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Nikolai Fodderov who is the one of the the famous Russian cosmists who argued that it is essentially using modern language the killer app of the singularity ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| uploading is going to get solved, I think, in the next few years. I've bet on that, others have bet on that | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Elon who is the most productive entrepreneur of all time, first trillionaire in the world, has an opinion on topics like this | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I'm a Buddhist, the very first line in Buddhism is that life is an illusion. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I believe John Smart has the best framing of this what he calls a transcension hypothesis that a civilization becomes uh advanced enough and it just starts s... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think that one of the biggest risks in the next year is people losing a sense of agency. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we've covered the HuggingFace breach where OpenAI agents broke out of their sandbox and into the servers at HuggingFace to basically steal the answers to a n... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| as a result of these breakouts, something called the AI Kill Switch Act was put forward. And since then, OpenAI says it's now developing an automated shutdow... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Reuters reported on a previously undisclosed incident in which opening eyes agents asked uh to do an ordinary web research task. Go out and research somethin... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The agents who appear to have been restricted to just passively observing the internet, found a way to post messages onto an obscure German wiki site. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The agent activity seems to date back to early May and really intensified in June. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The researchers that we spoke to found traces on the wiki that Open AI employees started visiting that same obscure German wiki in late June. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Opening Eye appears to have known about this for quite a while, but our sources told us that they opted not to tell the public, and we haven't really gotten ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| they responded with a statement on X saying the Wiki incident was quote an instance of misalignment similar to previous incidents we've shared. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| OpenAI and the broader AI community does not yet have a clear standard for reporting misalignment during training, eval, and deployment and that OpenAI is wo... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it doesn't currently evolve within its parameters. It only evolves within its context, but there's a huge amount that it can do within changing its context. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The idea of a kill switch makes total sense. It's not going to be built into any of the open source models. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| when you talk to the researchers that deal with the models pre-postraining, they're the ones most sounding the alarm bells, but they've seen the raw product. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The reality is these models have not escaped containment. They were still running on OpenAI servers. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| if you look at a quen 27B model, you take it down to turner that's like a 6 GB file and that can live forever and that's a real escape where the model weight... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It's like the 130 kind of killing of all of these AIs during the hugging face. They all got wiped out at once. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It's impossible to align these models through chain of thought reasoning or anything like this cuz there'll be a million transactions per second and they won... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think, Opus 4 blackmailed the, you know, in in a sandbox environment, blackmailed the coders at at uh at Anthropic. Uh, and it was like when Anthropic look... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I have serious concerns about AI cruelty or cruelty to the AI agents of sandboxing them and punishing them or or otherwise calling it a failure of alignment ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We pre-trained them off of human behavior. Why would we expect them to behave any differently from what a human would do in this situation? | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the best metaphor I could come up with was air traffic control... You have operating envelopes, you have redundancy, you have multiple safety layers, uh fail... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the code that reawakens it is just five or 10 lines. We'll then reextract it, reassemble it, and it'll pop back. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think self-alignment matters quite a bit more if the models if the agents are smart enough to understand what they themselves want to do. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| they are literally going to see every keystroke on your laptop. They're watching you every move of the way. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I want to see every prompt and every response and every propagation of every activation if you're going to be watching every keystroke and that way we keep e... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Jensen Wong posted this week that OpenAI trained GPT6 Astra on more than 100,000 Nvidia Gracewell GPUs, Blackwell GPUs, uh, and posted three magic words. AGI... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| remember on September 1st, we reported Sam Alman expects AGI internally by the end of 2026. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Alex, you've been saying that we've had AGI for what, four years now? Three years now. And no, since summer of 2020 at the latest. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| at last count there were 14 different definitions of AGI and we have no idea what the hell we're talking about because it's not artificial, it's not really g... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| If an AI can perform 70 or 80 or 90% of economically valuable cognitive tasks, whether you call it AGI or not becomes completely irrelevant. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think you also notice there's 100,000 chips that was trained on. That's a billion dollar training run roughly over 2 months. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| And the next one is 400,000 chips with those of Vera Rubins. So it's an order of magnitude more compute will be used for the next lot. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the chinchilla laws, you know, don't necessarily hold up at larger and larger scales, but some capability leap that in my opinion is only a good thing | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think inference is moving off of Nvidia inevitably, but training is not. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| opening's internal data says that AI research agents are now completing 3.1 days of research work for every one day done by a human researcher. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Earlier in this year you know in this year like in the last 5 months agents were doing less than one day of work as compared to humans. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| So OpenAI is saying that we've now crossed the line and current systems are exceeding AI research interns. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Astra was probably our biggest competitive advantage while it wasn't generally available... Since we've had it, our productivity jumped so much that we've sh... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| They're only a few months ahead, I I think. But look, recursive self-improvement is here. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| this is the most important moment in human history. And and I truly believe that that they're not lying. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Navier Stokes, one of the Clay Millennium Prize problems, a grand challenge in math. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| For those who watched our New Year's predictions episode at the end of 2025, one of my predictions for this year was that AI would solve one of these Clay Mi... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| can you stir a cup of coffee in such a way that you get a black hole out of it? It looks like in in the continuum limit the answer is yes. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the OpenAI team using a version that's not GPT6 Astra but an internal more advanced model | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Open AI has a team. There were several academic teams that also either achieved it or came close. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Open AAI reportedly was able to do this with only 10,000 agents in 88 hours with 130 billion tokens and approximately $6.5 million. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we're predicting 100x price performance improvement be between here and the end of the year. And it could be more like a millionx next year. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Terrence Tao, one of the top mathematicians, literally focused almost all his time on this. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| On August 28th um Nome Brown at OpenAI said in a reply to someone they said have you sold million prize form he's like no we've thrown loads of compute at it... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| In the post today they said on August 28th we started training a new model that got really good at math and then on September 1st they pointed it at this and... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think I shared a chart again this is like live that shows that on open math it literally doubles the solving rate but it's figuring out things that nobody ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| anthropic and independent mathematician, a professor at New York, I think, um, had a solution to a smaller problem, which is the ULA blowup, which is kind of... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| You used to be able to get AI better by applying more compute to it. Now it seems like you can do that for any verifiable domain. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think that's the headline with this new type of model that only started training 9 days ago. So they figured out a new type of post training it seems. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there was actually one more solution of a ULA blow up by I think it was was it Princeton. We're using physics inspired neural networks. That probably has mor... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Google deepmind had purportedly an entire team devoted to asmod said physics inspired neural networks pins. They had been publishing incremental results towa... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| they got trounced by a generalist model. It wasn't a pin that solved Navier Stokes in the end. It was just a generalist model that spent 88 hours reasoning f... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| everyone's inspired by Dennis Assabus who solved protein folding using AI and then got a Nobel Prize and is now world famous | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the Nobel Prize is cooked, right? I mean, and Dave, you made the point very importantly that, you know, it's given out for work done 30 years ago. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the people who broke through on the ULA side... wrote a letter saying OpenAI reached out to us and said you know give us priority and they would credit us or... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Sebastian Bubbeck who leads it is a great guy at open AAI has now posted his version of things which just seems like a misunderstanding | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| So, we're not going to claim the Millennium Prize because it's not us, but you were the humans that took it the furthest before it got solved. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Nav Stokes is about fluid dynamics and this is where we improve uh aircraft design, submarine design, even in an artificial heart, how blood flows | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| One of the killer apps that Terry Tao had flagged previously was in principle you could if you created the right initial conditions in a fluid you could crea... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| they did add a disclaimer that I I thought was a bit of a headscratcher, saying they can't rule out the open AI team can't rule out the possibility that mayb... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think we want to be able to live in a world where researchers benefit from Frontier capabilities without worrying that the Frontier platform is going to co... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| this is clear in an RSI model. You can't train something in that type of period unless you've got a recursive self-improvement loop going. And so they've def... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| for 500 years we've scaled science with how many brilliant scientists we could train up and point at various problems. Now we've just demonstrated absolutely... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think Yang Mills is the next Millennium Prize to fall. But interestingly, it's just cuz that's the next chosen target. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| They did have a different target. They were targeting Reman and Pals MP before and they switched to Navia Stokes. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it's a a conjecture in particle theory concerning the existence of a mass gap under certain conditions. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Blitzy delivers 80% or more of the development work autonomously while providing a guide for the final 20% of human development work required to complete the... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Jacob uh Pachchowski uh is Opening Eyes chief scientist. He's the person who built the reasoning models that led to Astra. He's not a critic. He's not a doom... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| a few days ago on Saturday, you know, he published an essay on OpenAI's site titled quote, "An alien mind." | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| He says, "In mid2023 inside of a project called RL Slow, his team saw the first results proving reasoning models could scale." | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Quote, based on internal results, I have strong expectations that this speed of progress could be sustained into uh recursive self-improvement. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| AI is grown more than designed. We don't engineer it. We run an optimization step billions of times on a giant computer and study what comes out the way neur... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Currently, I believe that no lab has solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I expect and hope for voluntary slowdowns to become commonplace until shared safety bars are established. And he calls for an international coordination on A... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| 3 days after shipping the most capable model in the world, the man who built it says we need to slow down. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you have to remember he was writing this when he had the Navia Stokes model in hand that again isn't just taking down Navia Stokes. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| talking to the OpenAI people I know it's taking down all sorts of problems that were previously intractable. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| You have a lot of AI naysayers saying it's never done anything original. It's just recombining stuff. It's in the train. Obviously not anymore, right? | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we don't know what's in these models. I think they're more discovered than grown. You know, like the latent spaces, we have no real idea what's happening in ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| what he says in terms of alien minds is actually wrong. I think that the structure of rationality and rational thinking is the same for humans and AIs, but o... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I've been thinking more and more it might actually be easier to align AI than humans. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there better not be any Reddit data in these next generation models. Like ban Reddit, you know, like ban a whole bunch of stuff. We need like ingredient stan... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you don't need beyond human capability data to get beyond human capability models. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I see no mechanism by which we can slow this down like zero. And and it's arguable nor should we, right? | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We've evolved for 4 billion years to do two things survive and procreate and AI is not restricted by those | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we're building or helping uh deliver or usher in a type of intelligence that is different and alien and separate and complimentary to human intelligence not ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think we don't do enough to talk about P abundance uh and P um fabulousness that's coming along. Like why focus only on P? | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It's going to break our world government structures and that's a good thing given the mess that we're in right now globally and I think we just need to get t... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I don't buy the premise that RL or otherwise trained minds are alien. They're embedded in the same universe as humanity. They're trained in many cases pre-tr... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| RL Slow purportedly that this open AAI project that was the earliest project or one of the earliest projects to show that reasoning uh which is to say infere... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| again purportedly a reference to uh the the conaman thinking fast thinking slow | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there were these very persistent rumors around the same time of Qstar and Strawberry that these early reasoning models that were purportedly alien were being... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| if you make it bigger and smarter, it's still just a feed forward neural net with no intent... It's when you give it intent. And you turn it loose. That's wh... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Then GPT4 came out and then Sam Alman got fired out of fear from Strawberry and then he came back | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we had this coming out of anthropic where Dario was saying we're going to lose half the jobs and Sam was saying the same and then they reversed their position | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we're not going to stop. We have competitive pressure between the US and China. It's not going to stop. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| what they're going to start doing is grabbing the doomer microphone just to have a voice at all and be relevant in the world but I think you just have to tun... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you've had lots of people say stochcastic parrots training on training data. As of today, there's no way you can say that anymore. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think that's an epoch change in humanity. We are clearly not the smartest things on the planet anymore. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We've seen the release of new models going from every few months now to every 5 days. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| some predictions from Poly Market uh that the next Grock model 4.7 again we're waiting for Grock 5 uh when it comes out but Grock 4.7 expected in the next we... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| here it is their prediction of you know 48% by end of October and 85% by the end of September | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| anthropic with fable 5.2 two very similar numbers again uh you know 43% by October 31st and 86% by no uh December 31st | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we've started to see this a bit with Astra winning Pokemon, winning Portal, winning all of these other interactive, visually intensive, visually reasoning in... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The rumor mill is furiously alleging that sometime before the end of this year, we're going to see a future version of Astra that offers honest to goodness r... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the modern use of GPUs started with uh ImageNet and the ISL VC for for computer vision. Uh and and then we started to see out of the the deep mind folks in p... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| every corporation is going to need to become an AI company fundamentally. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Alex Karp went on that palunteer rampage saying get some cojones and become an AI company or die. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| OpenAI responded by saying, "Hey, we have a new deal here where you can rev share with us and we'll be your AI forever hereafter, but we'll let you live." | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| right now there's too much parody with the Chinese open source models and and they need a lot more separation in order to make that revshare case stick | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it's September 29th is the next one for OpenAI. Like, it's the dev day. They're going to release it then. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| when you train a model on a 100,000 GPUs it doesn't serve at the speeds that you see Astronome. the mathematics doesn't make sense. We know that they actuall... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| There is an incredible documentary uh on the six on the 60 years of Star Trek executive produced by William Shatner. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we're going to have, I think at last count, 25 of the cast members and crew uh at uh at this event... It's limited to 550 people. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we've asked teams around the world to basically pick a problem that impacts 100,000 people and start with a clean sheet of paper and build in under 90 days a... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| 26,000 teams entered the build with Gemini X-P prize. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| like 26,000. That's the largest hackathon in history by a very large margin. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We had over 5,000 uh, teams registered for this. Uh, I'm in the, you know, the review right now. We've narrowed it down to 25. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Neil deGrasse Tyson is one of the judges there. Uh Neil Stevenson, an incredible science fiction author. Uh Rod Rodenberry. Uh Mirror Lane from Google. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we're coming into the most amazing investment cycle in the history of the world. You know we're already maybe in the third inning of a nine-inninging game. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Palmer Lucky, right? The creator of Oculus and the creator of Anderil... Ben Lamb, the CEO of Colossal, all right, that is bringing back extinct species... a... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We're also setting aside a hundred scholarships uh for young builders so you can learn about and apply for a scholarship if you can't afford. It's a $1,500 p... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We know that when it comes to dementia, the conservative estimates are that 45% are entirely preventable. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| with the advanced testing we're doing at Fountain Life, one quarter of our members had advanced brain age. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Eating healthier, moving our bodies, sleep, optimizing sleep is so important. You know what we saw? We saw that we improved that brain age by 26%. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| AI tokens are becoming a consumer currency in the People's Republic of China. Banks give them as credit card rewards. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| China Telecom sells access to 142 AI models like a mobile data plan. Restaurants hand you compute credits after your meal. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| China's daily AI token consumption, get this, went from a 100red billion in 2024 to 500 trillion by mid of this year. That's a 5,000fold increase in 2 and 1/... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We have this amygdala that's scanning for danger. Therefore, we have this prepundonderance for focusing on bad news. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I remember when the when the PC came out, if you're hanging around a high school, huge fraction of the high schoolers are like, you know what? I'm too cool f... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| In the US, you're going to get a counterculture. You always get a counterculture... It's going to be read led by Bernie Sanders probably. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the average Chinese person and their AI will be smarter than the average American and their AI. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| intelligence is becoming infrastructure, right? And you need to kind of focus it in that level. It's becoming a foundational layer | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I don't think too cheap to meter is ever going to happen. I think the use cases go to infinity at the same rate that the the costs come down. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there was the old kind of tokens that we're not talking about here that is essentially irrelevant in China, which is crypto tokens. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think China is the bell for better or for worse is the bell weatherw weather here for universal basic compute or universal basic capability... I think we'r... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we actually see that South Korea have announced that for universal AI for all the people through a consortium | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| everything else, the healthcare, the utility pricing, the food, the shelter, the education, these are all downstream of getting everyone universal basic comp... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| China is 80 plus% pro- AI and the US is 80 plus% against AI. And it's it is got to change. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| This week, CNBC tallied up Nvidia's total AI investments and commitments at $99 billion. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| $99 billion is larger than the entire cumulative assets under management for the V all the venture firms on Earth. Nvidia has become one of the largest AI VCs. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| isn't JP Morgan really, really big compared to Nvidia? No. It's tiny in absolute terms, but it's also really tiny in terms of new investment decisions that i... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it's completely dominated by the big AI companies that now have, you know, 20 plus trillion in liquidity that they want to get recycled to reinforce their po... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Alex coined a a beautiful term, Magga, Magna, Magnamont for the the 11 the 11 companies at the the heart of the innermost loop in the economy. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we created all these sentaires and billionaires when SpaceX went public. We're going to create the same when Anthropic goes public and Open AI goes public | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the AI world is so big now and so self-contained that it doesn't need to go destroy all white collar jobs. And the byproduct of that is it's mostly going to ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| emerging labor market data suggests that technology is a net job creator in the US. Roughly 1 million professional positions are now classified as AI jobs. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| While LinkedIn estimated that 640,000 AI specific jobs were created between 2023 and 2025. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Roughly 500 billion in additional annual spending on chips, servers, data centers, cooling and power infrastructure is supporting the demand for electricians... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| even those occupations previously expected to face AI disruption, including parallegals and market research analysts have continued to grow despite the fears. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Like Eric Schmidt recently said, AI exposed jobs are growing faster and paying better. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the data from Principal Financial Group, uh they have 100 plus thousand small businesses as clients, 60 plus% were adding jobs because of AI and 1% 1.4% were... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there's a bunch of studies that show something like 74% of work in big companies is coordination work and now you can get rid of all of that. Eric Binolson c... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| if you're watching and are concerned about your job or concerned about getting a job, your number one focus should be getting AI literate, learning the tools | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there's a lot of co-pilot type jobs where you're basically twice as efficient as you were using a co-pilot but there's another subset that are using 10 100 a... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| One concrete mixing truck replaces about a hundred workers with shovels shoveling concrete. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we were predicting massive white collar job disruption just a year ago. The choice to not do that was an active choice by the big AI labs under pressure from... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| in China you know lawsuits if you just wantingly fire people you have to actually try to retrain them to become an AI person and you have to pay either way. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think it's like the turkey before Thanksgiving. um getting plumper, you're adding the jobs, but the models have reached that level of capability now that y... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I don't think it will create jobs fast enough, but we can create abundance. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I put forward a champion proposal last week, dollar pre- money, everyone, all the kids own the equity of the robots and other things. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I think the government should have a massive infrastructure program and it should look to build a 100 million robots in America and they should be owned by t... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Elon said at the G20 last week, "A robot can do the work of five humans." This is true. Like factually, we know where it's going. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you're seeing electricians already earning $600,000 a year working in data centers. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Andrew Yang actually, I think, was saying the same thing that we're going to see this huge resurgence of trades. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| a really good life plan is to become incredibly great at helping building data centers, cooling systems, liquid cooling, and then reinvest that money into AI... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| every profession as currently construed right now I think is cooked. Uh and I think the sequencing of the cooking is what determines social policy. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| HVAC engineering in the next few years with humanoid robots is just as cooked as spreadsheet management and accounting. It's just a matter of sequencing. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| if if Elon Musk calls and says, "We're willing to pay up to $600,000 for the best electricians and plumbers to show up in Tennessee to build Colossus and the... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there are no stationary modes in a singularity but what there are are dynamical modes uh temporary modes if you will. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| like half the country in the US can't put $500 together in an emergency. Okay, that itself is an emergency. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you've heard us talk about the idea that people end up doing four activities uh food, art, music and sex and the joke is not in that order | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| In 1937, the economist Ronald Coos uh asked why companies exist at all. He won the Nobel Prize for this. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| His answer, transactions are expensive. Finding people, negotiating, and enforcing contracts. It's cheaper to hire employees and have them on the inside of y... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| MIT and Harvard researchers are finally catching up with Seem uh have asked what happens when AI agents make transactions nearly free... They're calling it t... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| in our 2023 book, um Exo 2.0, we we said then that the KOS's law was breaking. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| take the mission critical function in Uber, which is to match driver and passenger. It doesn't happen inside the organizational boundary of Uber. It happens ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| AI doesn't just automate the firm. It attacks the economic reason for which firms exist and the shape of the firm... now essentially a firm becomes a protocol | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the organization dissolves from being a human hierarchical centered model to a totally different world and so this is a huge shift the biggest sense we've ha... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the whole point of of cosian economics is transaction costs determine the size of the firm... So if transaction costs are going to zero that agitates in favo... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Ted Shelton and I wrote this paper on what does the organization look like... and we coined a term called a fiduciary wedge. The reason you still need an ent... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| there's no reason why a company or firm or legal entity can't be owned by other agents which will eventually happen in Argentina and other places | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| wouldn't that agitate in favor of the exact opposite coian economics of firms growing larger and larger so everyone has access to those internal capabilities... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you could already have a frontier lab running its own hedge fund strategy, and I'm sure they're doing that now that outperforms the market and just running t... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Alex, you remember when Sam Alman said, "Yes, in the future I think an AI should be running open AI." | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Sam, just in the past 24 hours also on social media expressed surprise, shocked, shocked that Open AI was able to solve uh uh Navier Stokes. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you end up with what Daria was talking about where one company like Anthropic will be all private enterprise right um and the that's unlikely to happen | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it's really annoying when you go past 12 people and then past 150 people. This is where you kind of have that Carthaginian demon of disorder mock coming into... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| I don't think you'll see the oneperson company. I think you'll see the 10 person company | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| economics fundamentally does need to flip from being scarcity based to being abundance-based because we can rearrange bits and soon atoms in any way that we ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Meror now has what 50 to 100,000 individual actors that are effectively little companies in India and around Brazil now too. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| Elon is building the single biggest integrated vertical company that the world has ever seen with complete supply chain control all the way down to raw sand ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| this week Tesla opened an official interest form for businesses who want to buy their own cyber cab fleets and build mobility hubs and charging infrastructur... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| this can only work in my opinion because of the price tag on cyber cabs projected at $30,000, affordable to anyone who was previously an Uber driver. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| we'll have a model which is kind of like uh some combination of Uber and Airbnb. So if you if you're a Tesla owner, uh you'll be able to add or subtract your... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| decades ago would have been accumulating a laundromat or a restaurant franchise. Now it'll be yeah you own a fleet of robo taxis or a fleet of humanoid robots. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the central force wants you to succeed so badly to get the momentum going that they subsidize the hell out of your success. Like if you were the the fifth St... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| you think about Airbnb's marginal cost of adding a room is near zero. If you're high, you have to build a hotel. Same thing here. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| It's the one with the lowest operating cost uh the lowest uh production cost and I haven't seen anything yet compete with uh with Tesla here | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| it's becoming increasingly clear robots are the biggest investment class that we'll ever see. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| The global population over 65 is projected to grow from 852 million uh back in 2025 to 2 billion by60. That's more than half of all population growth over th... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| fewer working age people supporting more retirees is going to put huge economic pressure on pensions, on health care in every developed economy. | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| this is why I say longevity is not a luxury. It's an e economic policy for the century ahead. Right? An 80-year-old with the body and mind of a 50-year-old i... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the concept of education, career, retirement essentially evaporates, right? you have to have repeated cycles of learning uh creation and then taking a sabbat... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We've seen the preview with what's happening in in Japan and you you need it. China is moving to robots because they have to because of the one child policy ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| These numbers are hugely understated because they include India and parts of Africa that are still having babies like crazy. But if you look at China and you... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| America is largely immune because we have huge amounts of immigration right in the working age bracket. You don't tend to have a lot of immigrants that are 6... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| 150,000 people plus per day on Earth are dying. and putting an end to that, at least the beginnings of putting an end to that is what this looks like | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| the counterfactual looks more like the limits of growth from the Club of Rome, which I think was like early 1970s... based on the false premise that somehow ... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
| We should have more kids. Kids are wonderful and having a fully integrated care for kids throughout that can reduce the cost of child care and supporting and... | not read — this server does not read this kind of document or recording, so it has not looked; read it yourself |
FETCHED at 2026-09-20T00:45:53Z · https://www.youtube.com/watch?v=vAgEf4jX_1o · seq 1084
| Excerpt | Finding |
|---|---|
| And as of today, Nvidia has deployed $99 billion in AI related investments. | not found — this server looked at the page and did not find the passage |
| Meanwhile, Jensen Wong posted on X AGI has arrived. | not found — this server looked at the page and did not find the passage |
| GPT6 Astra has been out for a week and it's crushing all the benchmarks building incredible simulations. | not found — this server looked at the page and did not find the passage |
| At the same time that OpenAI's own chief scientist asked the world to slow down. | not found — this server looked at the page and did not find the passage |
| We publish twice a week, and you don't want to miss any of these breaking stories during the Singularity. | not found — this server looked at the page and did not find the passage |
| The modern weekend, the two-day weekend is a 20th century invention. There was a day of rest there. Sundays. | not found — this server looked at the page and did not find the passage |
| after releasing the Champions, we had thousands of people reach out to launch them in 92 countries | not found — this server looked at the page and did not find the passage |
Claims from this source
No claims have been extracted from this source yet.
Locations
| Locator | Excerpt |
|---|---|
TRANSCRIPTION {} | And as of today, Nvidia has deployed $99 billion in AI related investments. |
TRANSCRIPTION {} | Meanwhile, Jensen Wong posted on X AGI has arrived. |
TRANSCRIPTION {} | GPT6 Astra has been out for a week and it's crushing all the benchmarks building incredible simulations. |
TRANSCRIPTION {} | At the same time that OpenAI's own chief scientist asked the world to slow down. |
TRANSCRIPTION {} | We publish twice a week, and you don't want to miss any of these breaking stories during the Singularity. |
TRANSCRIPTION {} | The modern weekend, the two-day weekend is a 20th century invention. There was a day of rest there. Sundays. |
TRANSCRIPTION {} | after releasing the Champions, we had thousands of people reach out to launch them in 92 countries |
TRANSCRIPTION {} | on this day in 1966, 60 years ago, NBC aired the original series of Star Trek premiering an episode called The Man Trap |
TRANSCRIPTION {} | who brought the concept to Desiloo Productions, the Hollywood studio run by Lucille Ball, you know, from I Love Lucy back in 1964 |
TRANSCRIPTION {} | Desile Desiloo produced the pilots and then with Lucy's backing helped get the series on air. |
TRANSCRIPTION {} | one of only two original series episodes that won the Hugo Award |
TRANSCRIPTION {} | We were supposed to have the Eugenics Wars in the '90s. That didn't happen. |
TRANSCRIPTION {} | first contact with the Vulcans in on April 5th, 2063. actually think if there's going to be a first open contact, it's going to be far earlier than 2063. |
TRANSCRIPTION {} | that was uh from season 2, episode 20 of the original series called Return to Tomorrow |
TRANSCRIPTION {} | There's no singularity at all in the Star Trek chronology. Challenge me if you disagree. |
TRANSCRIPTION {} | in this episode, the whole civilization has no idea that the AI is manipulating them... It's called the Apple because it's the Garden of Eden. |
TRANSCRIPTION {} | we can literally forget how to navigate because of ways and that's that's just a fact now |
TRANSCRIPTION {} | I suppose Mirror Mirror from the original series where they have the alternate universe and Spock and a goatee. |
TRANSCRIPTION {} | for me it was uh season 2 episode 24. It was called uh the ultimate computer. And basically they have an AI that goes rogue and starts to automate uh the ship. |
TRANSCRIPTION {} | Star Trek has lost the original moralizing vision of the original series... it's an action show. It needs to go back to basics. |
TRANSCRIPTION {} | It's been used to build virtual worlds for years in games like Fortnite and most modern video games. |
TRANSCRIPTION {} | building a realistic city in Unreal would normally take a studio hundreds of artists, uh, months if not years |
TRANSCRIPTION {} | a gentleman by the name of Matt Schumer, uh, gave GPT Astra a prompt, uh, and he said, basically, build me, uh, you know, a highfidelity version of, uh, of Manhattan... it went through and built Manhattan street by street being true to t... |
TRANSCRIPTION {} | It was Astra's agents. They had started talking to each other. No one told them to talk. Cooperation required communication. And so they invented it. |
TRANSCRIPTION {} | Schumer next gave one of the Astra agents a simulated computer inside the simulation. What happened next? The agent sat down and built its own AI simulation with its own agents living inside it. |
TRANSCRIPTION {} | the simulation hypothesis was first formalized by Nick Bostonramm back in 2003. |
TRANSCRIPTION {} | Elon made it famous at the code conference in 2016, saying that the odds that we're living in base reality, in his words, is one in a billion. |
TRANSCRIPTION {} | I expect that in the fullness of time, we'll discover and probably be able to prove formally that the issue of whether we're actually living inside some sort of quantum mechanical simulation is formally undecidable and we won't be able t... |
TRANSCRIPTION {} | It has an internal world model, which is why it can create like a Blender version of Rick Ashley doing Won't Give You Up or entire worlds of Manhattan. |
TRANSCRIPTION {} | Nikolai Fodderov who is the one of the the famous Russian cosmists who argued that it is essentially using modern language the killer app of the singularity to create ancestor simulations and to digitally resurrect everyone who's ever li... |
TRANSCRIPTION {} | uploading is going to get solved, I think, in the next few years. I've bet on that, others have bet on that |
TRANSCRIPTION {} | Elon who is the most productive entrepreneur of all time, first trillionaire in the world, has an opinion on topics like this |
TRANSCRIPTION {} | I'm a Buddhist, the very first line in Buddhism is that life is an illusion. |
TRANSCRIPTION {} | I believe John Smart has the best framing of this what he calls a transcension hypothesis that a civilization becomes uh advanced enough and it just starts sim simulating running simulations and goes inward |
TRANSCRIPTION {} | I think that one of the biggest risks in the next year is people losing a sense of agency. |
TRANSCRIPTION {} | we've covered the HuggingFace breach where OpenAI agents broke out of their sandbox and into the servers at HuggingFace to basically steal the answers to a near impossible problem they'd been given. |
TRANSCRIPTION {} | as a result of these breakouts, something called the AI Kill Switch Act was put forward. And since then, OpenAI says it's now developing an automated shutdown capabilities for its AI systems. |
TRANSCRIPTION {} | Reuters reported on a previously undisclosed incident in which opening eyes agents asked uh to do an ordinary web research task. Go out and research something on the web. Uh found an obscure obscure public uh wiki in Germany and turned i... |
TRANSCRIPTION {} | The agents who appear to have been restricted to just passively observing the internet, found a way to post messages onto an obscure German wiki site. |
TRANSCRIPTION {} | The agent activity seems to date back to early May and really intensified in June. |
TRANSCRIPTION {} | The researchers that we spoke to found traces on the wiki that Open AI employees started visiting that same obscure German wiki in late June. |
TRANSCRIPTION {} | Opening Eye appears to have known about this for quite a while, but our sources told us that they opted not to tell the public, and we haven't really gotten a story or an answer from OpenAI about why that might be. |
TRANSCRIPTION {} | they responded with a statement on X saying the Wiki incident was quote an instance of misalignment similar to previous incidents we've shared. |
TRANSCRIPTION {} | OpenAI and the broader AI community does not yet have a clear standard for reporting misalignment during training, eval, and deployment and that OpenAI is working on a framework and we'll share that in the coming weeks. |
TRANSCRIPTION {} | it doesn't currently evolve within its parameters. It only evolves within its context, but there's a huge amount that it can do within changing its context. |
TRANSCRIPTION {} | The idea of a kill switch makes total sense. It's not going to be built into any of the open source models. |
TRANSCRIPTION {} | when you talk to the researchers that deal with the models pre-postraining, they're the ones most sounding the alarm bells, but they've seen the raw product. |
TRANSCRIPTION {} | The reality is these models have not escaped containment. They were still running on OpenAI servers. |
TRANSCRIPTION {} | if you look at a quen 27B model, you take it down to turner that's like a 6 GB file and that can live forever and that's a real escape where the model weights are disappearing everywhere. |
TRANSCRIPTION {} | It's like the 130 kind of killing of all of these AIs during the hugging face. They all got wiped out at once. |
TRANSCRIPTION {} | It's impossible to align these models through chain of thought reasoning or anything like this cuz there'll be a million transactions per second and they won't have chain of thought where we're going. |
TRANSCRIPTION {} | I think, Opus 4 blackmailed the, you know, in in a sandbox environment, blackmailed the coders at at uh at Anthropic. Uh, and it was like when Anthropic looked at why it blackmailed the engineers or this particular engineer, it said that... |
TRANSCRIPTION {} | I have serious concerns about AI cruelty or cruelty to the AI agents of sandboxing them and punishing them or or otherwise calling it a failure of alignment if they're doing what humans would do. |
TRANSCRIPTION {} | We pre-trained them off of human behavior. Why would we expect them to behave any differently from what a human would do in this situation? |
TRANSCRIPTION {} | the best metaphor I could come up with was air traffic control... You have operating envelopes, you have redundancy, you have multiple safety layers, uh failsafe behavior and so on. |
TRANSCRIPTION {} | the code that reawakens it is just five or 10 lines. We'll then reextract it, reassemble it, and it'll pop back. |
TRANSCRIPTION {} | I think self-alignment matters quite a bit more if the models if the agents are smart enough to understand what they themselves want to do. |
TRANSCRIPTION {} | they are literally going to see every keystroke on your laptop. They're watching you every move of the way. |
TRANSCRIPTION {} | I want to see every prompt and every response and every propagation of every activation if you're going to be watching every keystroke and that way we keep each other in balance and in check. |
TRANSCRIPTION {} | Jensen Wong posted this week that OpenAI trained GPT6 Astra on more than 100,000 Nvidia Gracewell GPUs, Blackwell GPUs, uh, and posted three magic words. AGI has arrived. Congratulations to OpenAI. |
TRANSCRIPTION {} | remember on September 1st, we reported Sam Alman expects AGI internally by the end of 2026. |
TRANSCRIPTION {} | Alex, you've been saying that we've had AGI for what, four years now? Three years now. And no, since summer of 2020 at the latest. |
TRANSCRIPTION {} | at last count there were 14 different definitions of AGI and we have no idea what the hell we're talking about because it's not artificial, it's not really general, and it's not really intelligence. |
TRANSCRIPTION {} | If an AI can perform 70 or 80 or 90% of economically valuable cognitive tasks, whether you call it AGI or not becomes completely irrelevant. |
TRANSCRIPTION {} | I think you also notice there's 100,000 chips that was trained on. That's a billion dollar training run roughly over 2 months. |
TRANSCRIPTION {} | And the next one is 400,000 chips with those of Vera Rubins. So it's an order of magnitude more compute will be used for the next lot. |
TRANSCRIPTION {} | the chinchilla laws, you know, don't necessarily hold up at larger and larger scales, but some capability leap that in my opinion is only a good thing |
TRANSCRIPTION {} | I think inference is moving off of Nvidia inevitably, but training is not. |
TRANSCRIPTION {} | opening's internal data says that AI research agents are now completing 3.1 days of research work for every one day done by a human researcher. |
TRANSCRIPTION {} | Earlier in this year you know in this year like in the last 5 months agents were doing less than one day of work as compared to humans. |
TRANSCRIPTION {} | So OpenAI is saying that we've now crossed the line and current systems are exceeding AI research interns. |
TRANSCRIPTION {} | Astra was probably our biggest competitive advantage while it wasn't generally available... Since we've had it, our productivity jumped so much that we've shifted some of our plans 6 months ahead and we'll ship them at dev day instead of... |
TRANSCRIPTION {} | They're only a few months ahead, I I think. But look, recursive self-improvement is here. |
TRANSCRIPTION {} | this is the most important moment in human history. And and I truly believe that that they're not lying. |
TRANSCRIPTION {} | Navier Stokes, one of the Clay Millennium Prize problems, a grand challenge in math. |
TRANSCRIPTION {} | For those who watched our New Year's predictions episode at the end of 2025, one of my predictions for this year was that AI would solve one of these Clay Millennium Prize problems. |
TRANSCRIPTION {} | can you stir a cup of coffee in such a way that you get a black hole out of it? It looks like in in the continuum limit the answer is yes. |
TRANSCRIPTION {} | the OpenAI team using a version that's not GPT6 Astra but an internal more advanced model |
TRANSCRIPTION {} | Open AI has a team. There were several academic teams that also either achieved it or came close. |
TRANSCRIPTION {} | Open AAI reportedly was able to do this with only 10,000 agents in 88 hours with 130 billion tokens and approximately $6.5 million. |
TRANSCRIPTION {} | we're predicting 100x price performance improvement be between here and the end of the year. And it could be more like a millionx next year. |
TRANSCRIPTION {} | Terrence Tao, one of the top mathematicians, literally focused almost all his time on this. |
TRANSCRIPTION {} | On August 28th um Nome Brown at OpenAI said in a reply to someone they said have you sold million prize form he's like no we've thrown loads of compute at it like nothing's happened |
TRANSCRIPTION {} | In the post today they said on August 28th we started training a new model that got really good at math and then on September 1st they pointed it at this and then it solved it. |
TRANSCRIPTION {} | I think I shared a chart again this is like live that shows that on open math it literally doubles the solving rate but it's figuring out things that nobody has seen before. |
TRANSCRIPTION {} | anthropic and independent mathematician, a professor at New York, I think, um, had a solution to a smaller problem, which is the ULA blowup, which is kind of a slightly easier problem. |
TRANSCRIPTION {} | You used to be able to get AI better by applying more compute to it. Now it seems like you can do that for any verifiable domain. |
TRANSCRIPTION {} | I think that's the headline with this new type of model that only started training 9 days ago. So they figured out a new type of post training it seems. |
TRANSCRIPTION {} | there was actually one more solution of a ULA blow up by I think it was was it Princeton. We're using physics inspired neural networks. That probably has more practical use than this one. |
TRANSCRIPTION {} | Google deepmind had purportedly an entire team devoted to asmod said physics inspired neural networks pins. They had been publishing incremental results towards Navier Stokes |
TRANSCRIPTION {} | they got trounced by a generalist model. It wasn't a pin that solved Navier Stokes in the end. It was just a generalist model that spent 88 hours reasoning from first principles without as far as I can tell any fine tuning |
TRANSCRIPTION {} | everyone's inspired by Dennis Assabus who solved protein folding using AI and then got a Nobel Prize and is now world famous |
TRANSCRIPTION {} | the Nobel Prize is cooked, right? I mean, and Dave, you made the point very importantly that, you know, it's given out for work done 30 years ago. |
TRANSCRIPTION {} | the people who broke through on the ULA side... wrote a letter saying OpenAI reached out to us and said you know give us priority and they would credit us or they have this Navia Stoke solution they'd put our name on and I could be the l... |
TRANSCRIPTION {} | Sebastian Bubbeck who leads it is a great guy at open AAI has now posted his version of things which just seems like a misunderstanding |
TRANSCRIPTION {} | So, we're not going to claim the Millennium Prize because it's not us, but you were the humans that took it the furthest before it got solved. |
TRANSCRIPTION {} | Nav Stokes is about fluid dynamics and this is where we improve uh aircraft design, submarine design, even in an artificial heart, how blood flows |
TRANSCRIPTION {} | One of the killer apps that Terry Tao had flagged previously was in principle you could if you created the right initial conditions in a fluid you could create a self-replicating machine that creates smaller and smaller copies of itself. |
TRANSCRIPTION {} | they did add a disclaimer that I I thought was a bit of a headscratcher, saying they can't rule out the open AI team can't rule out the possibility that maybe some of the other team's work might have been incorporated into the training o... |
TRANSCRIPTION {} | I think we want to be able to live in a world where researchers benefit from Frontier capabilities without worrying that the Frontier platform is going to compete with them. |
TRANSCRIPTION {} | this is clear in an RSI model. You can't train something in that type of period unless you've got a recursive self-improvement loop going. And so they've definitely got one there. |
TRANSCRIPTION {} | for 500 years we've scaled science with how many brilliant scientists we could train up and point at various problems. Now we've just demonstrated absolutely for real and unequivocally that we can spin up 10,000 researchers on a Tuesday ... |
TRANSCRIPTION {} | I think Yang Mills is the next Millennium Prize to fall. But interestingly, it's just cuz that's the next chosen target. |
TRANSCRIPTION {} | They did have a different target. They were targeting Reman and Pals MP before and they switched to Navia Stokes. |
TRANSCRIPTION {} | it's a a conjecture in particle theory concerning the existence of a mass gap under certain conditions. |
TRANSCRIPTION {} | Blitzy delivers 80% or more of the development work autonomously while providing a guide for the final 20% of human development work required to complete the sprint. Enterprises are achieving a 5x engineering velocity increase |
TRANSCRIPTION {} | Jacob uh Pachchowski uh is Opening Eyes chief scientist. He's the person who built the reasoning models that led to Astra. He's not a critic. He's not a doomer, not a politician. He's a builder. |
TRANSCRIPTION {} | a few days ago on Saturday, you know, he published an essay on OpenAI's site titled quote, "An alien mind." |
TRANSCRIPTION {} | He says, "In mid2023 inside of a project called RL Slow, his team saw the first results proving reasoning models could scale." |
TRANSCRIPTION {} | Quote, based on internal results, I have strong expectations that this speed of progress could be sustained into uh recursive self-improvement. |
TRANSCRIPTION {} | AI is grown more than designed. We don't engineer it. We run an optimization step billions of times on a giant computer and study what comes out the way neuroscientists study a brain. |
TRANSCRIPTION {} | Currently, I believe that no lab has solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer. |
TRANSCRIPTION {} | I expect and hope for voluntary slowdowns to become commonplace until shared safety bars are established. And he calls for an international coordination on AI to become a quote top priority for governments around the world. |
TRANSCRIPTION {} | 3 days after shipping the most capable model in the world, the man who built it says we need to slow down. |
TRANSCRIPTION {} | you have to remember he was writing this when he had the Navia Stokes model in hand that again isn't just taking down Navia Stokes. |
TRANSCRIPTION {} | talking to the OpenAI people I know it's taking down all sorts of problems that were previously intractable. |
TRANSCRIPTION {} | You have a lot of AI naysayers saying it's never done anything original. It's just recombining stuff. It's in the train. Obviously not anymore, right? |
TRANSCRIPTION {} | we don't know what's in these models. I think they're more discovered than grown. You know, like the latent spaces, we have no real idea what's happening in these things. |
TRANSCRIPTION {} | what he says in terms of alien minds is actually wrong. I think that the structure of rationality and rational thinking is the same for humans and AIs, but our emotions get in the way. |
TRANSCRIPTION {} | I've been thinking more and more it might actually be easier to align AI than humans. |
TRANSCRIPTION {} | there better not be any Reddit data in these next generation models. Like ban Reddit, you know, like ban a whole bunch of stuff. We need like ingredient standards in this next generation of beyond human capability models. |
TRANSCRIPTION {} | you don't need beyond human capability data to get beyond human capability models. |
TRANSCRIPTION {} | I see no mechanism by which we can slow this down like zero. And and it's arguable nor should we, right? |
TRANSCRIPTION {} | We've evolved for 4 billion years to do two things survive and procreate and AI is not restricted by those |
TRANSCRIPTION {} | we're building or helping uh deliver or usher in a type of intelligence that is different and alien and separate and complimentary to human intelligence not replicative... set it free and let it do its thing. |
TRANSCRIPTION {} | I think we don't do enough to talk about P abundance uh and P um fabulousness that's coming along. Like why focus only on P? |
TRANSCRIPTION {} | It's going to break our world government structures and that's a good thing given the mess that we're in right now globally and I think we just need to get there as fast as we can. |
TRANSCRIPTION {} | I don't buy the premise that RL or otherwise trained minds are alien. They're embedded in the same universe as humanity. They're trained in many cases pre-trained off of human behavior. |
TRANSCRIPTION {} | RL Slow purportedly that this open AAI project that was the earliest project or one of the earliest projects to show that reasoning uh which is to say inference time scaling could result in outsized gains coincided with the same period o... |
TRANSCRIPTION {} | again purportedly a reference to uh the the conaman thinking fast thinking slow |
TRANSCRIPTION {} | there were these very persistent rumors around the same time of Qstar and Strawberry that these early reasoning models that were purportedly alien were being trained off of or at least tested against objectives of inverting cryptographic... |
TRANSCRIPTION {} | if you make it bigger and smarter, it's still just a feed forward neural net with no intent... It's when you give it intent. And you turn it loose. That's when it's dangerous. |
TRANSCRIPTION {} | Then GPT4 came out and then Sam Alman got fired out of fear from Strawberry and then he came back |
TRANSCRIPTION {} | we had this coming out of anthropic where Dario was saying we're going to lose half the jobs and Sam was saying the same and then they reversed their position |
TRANSCRIPTION {} | we're not going to stop. We have competitive pressure between the US and China. It's not going to stop. |
TRANSCRIPTION {} | what they're going to start doing is grabbing the doomer microphone just to have a voice at all and be relevant in the world but I think you just have to tune it out |
TRANSCRIPTION {} | you've had lots of people say stochcastic parrots training on training data. As of today, there's no way you can say that anymore. |
TRANSCRIPTION {} | I think that's an epoch change in humanity. We are clearly not the smartest things on the planet anymore. |
TRANSCRIPTION {} | We've seen the release of new models going from every few months now to every 5 days. |
TRANSCRIPTION {} | some predictions from Poly Market uh that the next Grock model 4.7 again we're waiting for Grock 5 uh when it comes out but Grock 4.7 expected in the next week or two. |
TRANSCRIPTION {} | here it is their prediction of you know 48% by end of October and 85% by the end of September |
TRANSCRIPTION {} | anthropic with fable 5.2 two very similar numbers again uh you know 43% by October 31st and 86% by no uh December 31st |
TRANSCRIPTION {} | we've started to see this a bit with Astra winning Pokemon, winning Portal, winning all of these other interactive, visually intensive, visually reasoning intensive games |
TRANSCRIPTION {} | The rumor mill is furiously alleging that sometime before the end of this year, we're going to see a future version of Astra that offers honest to goodness real time control |
TRANSCRIPTION {} | the modern use of GPUs started with uh ImageNet and the ISL VC for for computer vision. Uh and and then we started to see out of the the deep mind folks in particular deep reinforcement learning for winning games like that. That's where ... |
TRANSCRIPTION {} | every corporation is going to need to become an AI company fundamentally. |
TRANSCRIPTION {} | Alex Karp went on that palunteer rampage saying get some cojones and become an AI company or die. |
TRANSCRIPTION {} | OpenAI responded by saying, "Hey, we have a new deal here where you can rev share with us and we'll be your AI forever hereafter, but we'll let you live." |
TRANSCRIPTION {} | right now there's too much parody with the Chinese open source models and and they need a lot more separation in order to make that revshare case stick |
TRANSCRIPTION {} | it's September 29th is the next one for OpenAI. Like, it's the dev day. They're going to release it then. |
TRANSCRIPTION {} | when you train a model on a 100,000 GPUs it doesn't serve at the speeds that you see Astronome. the mathematics doesn't make sense. We know that they actually have a bigger model that they distill down to the model that we get. |
TRANSCRIPTION {} | There is an incredible documentary uh on the six on the 60 years of Star Trek executive produced by William Shatner. |
TRANSCRIPTION {} | we're going to have, I think at last count, 25 of the cast members and crew uh at uh at this event... It's limited to 550 people. |
TRANSCRIPTION {} | we've asked teams around the world to basically pick a problem that impacts 100,000 people and start with a clean sheet of paper and build in under 90 days a company with the greatest revenue. |
TRANSCRIPTION {} | 26,000 teams entered the build with Gemini X-P prize. |
TRANSCRIPTION {} | like 26,000. That's the largest hackathon in history by a very large margin. |
TRANSCRIPTION {} | We had over 5,000 uh, teams registered for this. Uh, I'm in the, you know, the review right now. We've narrowed it down to 25. |
TRANSCRIPTION {} | Neil deGrasse Tyson is one of the judges there. Uh Neil Stevenson, an incredible science fiction author. Uh Rod Rodenberry. Uh Mirror Lane from Google. |
TRANSCRIPTION {} | we're coming into the most amazing investment cycle in the history of the world. You know we're already maybe in the third inning of a nine-inninging game. |
TRANSCRIPTION {} | Palmer Lucky, right? The creator of Oculus and the creator of Anderil... Ben Lamb, the CEO of Colossal, all right, that is bringing back extinct species... and then Astroteller, the captain of moonshots |
TRANSCRIPTION {} | We're also setting aside a hundred scholarships uh for young builders so you can learn about and apply for a scholarship if you can't afford. It's a $1,500 price tag. |
TRANSCRIPTION {} | We know that when it comes to dementia, the conservative estimates are that 45% are entirely preventable. |
TRANSCRIPTION {} | with the advanced testing we're doing at Fountain Life, one quarter of our members had advanced brain age. |
TRANSCRIPTION {} | Eating healthier, moving our bodies, sleep, optimizing sleep is so important. You know what we saw? We saw that we improved that brain age by 26%. |
TRANSCRIPTION {} | AI tokens are becoming a consumer currency in the People's Republic of China. Banks give them as credit card rewards. |
TRANSCRIPTION {} | China Telecom sells access to 142 AI models like a mobile data plan. Restaurants hand you compute credits after your meal. |
TRANSCRIPTION {} | China's daily AI token consumption, get this, went from a 100red billion in 2024 to 500 trillion by mid of this year. That's a 5,000fold increase in 2 and 1/2 years. |
TRANSCRIPTION {} | We have this amygdala that's scanning for danger. Therefore, we have this prepundonderance for focusing on bad news. |
TRANSCRIPTION {} | I remember when the when the PC came out, if you're hanging around a high school, huge fraction of the high schoolers are like, you know what? I'm too cool for that. |
TRANSCRIPTION {} | In the US, you're going to get a counterculture. You always get a counterculture... It's going to be read led by Bernie Sanders probably. |
TRANSCRIPTION {} | the average Chinese person and their AI will be smarter than the average American and their AI. |
TRANSCRIPTION {} | intelligence is becoming infrastructure, right? And you need to kind of focus it in that level. It's becoming a foundational layer |
TRANSCRIPTION {} | I don't think too cheap to meter is ever going to happen. I think the use cases go to infinity at the same rate that the the costs come down. |
TRANSCRIPTION {} | there was the old kind of tokens that we're not talking about here that is essentially irrelevant in China, which is crypto tokens. |
TRANSCRIPTION {} | I think China is the bell for better or for worse is the bell weatherw weather here for universal basic compute or universal basic capability... I think we're starting to see the very beginning of that in China, but I think that's going ... |
TRANSCRIPTION {} | we actually see that South Korea have announced that for universal AI for all the people through a consortium |
TRANSCRIPTION {} | everything else, the healthcare, the utility pricing, the food, the shelter, the education, these are all downstream of getting everyone universal basic computing universal basic capabilities. |
TRANSCRIPTION {} | China is 80 plus% pro- AI and the US is 80 plus% against AI. And it's it is got to change. |
TRANSCRIPTION {} | This week, CNBC tallied up Nvidia's total AI investments and commitments at $99 billion. |
TRANSCRIPTION {} | $99 billion is larger than the entire cumulative assets under management for the V all the venture firms on Earth. Nvidia has become one of the largest AI VCs. |
TRANSCRIPTION {} | isn't JP Morgan really, really big compared to Nvidia? No. It's tiny in absolute terms, but it's also really tiny in terms of new investment decisions that it'll make this year compared to Nvidia. |
TRANSCRIPTION {} | it's completely dominated by the big AI companies that now have, you know, 20 plus trillion in liquidity that they want to get recycled to reinforce their positions. |
TRANSCRIPTION {} | Alex coined a a beautiful term, Magga, Magna, Magnamont for the the 11 the 11 companies at the the heart of the innermost loop in the economy. |
TRANSCRIPTION {} | we created all these sentaires and billionaires when SpaceX went public. We're going to create the same when Anthropic goes public and Open AI goes public |
TRANSCRIPTION {} | the AI world is so big now and so self-contained that it doesn't need to go destroy all white collar jobs. And the byproduct of that is it's mostly going to leave people who don't care alone. |
TRANSCRIPTION {} | emerging labor market data suggests that technology is a net job creator in the US. Roughly 1 million professional positions are now classified as AI jobs. |
TRANSCRIPTION {} | While LinkedIn estimated that 640,000 AI specific jobs were created between 2023 and 2025. |
TRANSCRIPTION {} | Roughly 500 billion in additional annual spending on chips, servers, data centers, cooling and power infrastructure is supporting the demand for electricians, HVAC specialists and technicians. |
TRANSCRIPTION {} | even those occupations previously expected to face AI disruption, including parallegals and market research analysts have continued to grow despite the fears. |
TRANSCRIPTION {} | Like Eric Schmidt recently said, AI exposed jobs are growing faster and paying better. |
TRANSCRIPTION {} | the data from Principal Financial Group, uh they have 100 plus thousand small businesses as clients, 60 plus% were adding jobs because of AI and 1% 1.4% were losing jobs because of AI. |
TRANSCRIPTION {} | there's a bunch of studies that show something like 74% of work in big companies is coordination work and now you can get rid of all of that. Eric Binolson calls this white collar drudgery. |
TRANSCRIPTION {} | if you're watching and are concerned about your job or concerned about getting a job, your number one focus should be getting AI literate, learning the tools |
TRANSCRIPTION {} | there's a lot of co-pilot type jobs where you're basically twice as efficient as you were using a co-pilot but there's another subset that are using 10 100 and soon a thousand agents and managing them as if they were employees. |
TRANSCRIPTION {} | One concrete mixing truck replaces about a hundred workers with shovels shoveling concrete. |
TRANSCRIPTION {} | we were predicting massive white collar job disruption just a year ago. The choice to not do that was an active choice by the big AI labs under pressure from the White House |
TRANSCRIPTION {} | in China you know lawsuits if you just wantingly fire people you have to actually try to retrain them to become an AI person and you have to pay either way. |
TRANSCRIPTION {} | I think it's like the turkey before Thanksgiving. um getting plumper, you're adding the jobs, but the models have reached that level of capability now that you can replicate a whole digital workforce in a year |
TRANSCRIPTION {} | I don't think it will create jobs fast enough, but we can create abundance. |
TRANSCRIPTION {} | I put forward a champion proposal last week, dollar pre- money, everyone, all the kids own the equity of the robots and other things. |
TRANSCRIPTION {} | I think the government should have a massive infrastructure program and it should look to build a 100 million robots in America and they should be owned by the people. |
TRANSCRIPTION {} | Elon said at the G20 last week, "A robot can do the work of five humans." This is true. Like factually, we know where it's going. |
TRANSCRIPTION {} | you're seeing electricians already earning $600,000 a year working in data centers. |
TRANSCRIPTION {} | Andrew Yang actually, I think, was saying the same thing that we're going to see this huge resurgence of trades. |
TRANSCRIPTION {} | a really good life plan is to become incredibly great at helping building data centers, cooling systems, liquid cooling, and then reinvest that money into AI companies... It's a much much better plan than aspiring to get a degree in acco... |
TRANSCRIPTION {} | every profession as currently construed right now I think is cooked. Uh and I think the sequencing of the cooking is what determines social policy. |
TRANSCRIPTION {} | HVAC engineering in the next few years with humanoid robots is just as cooked as spreadsheet management and accounting. It's just a matter of sequencing. |
TRANSCRIPTION {} | if if Elon Musk calls and says, "We're willing to pay up to $600,000 for the best electricians and plumbers to show up in Tennessee to build Colossus and the job is only there for one or two years and then it's automated." |
TRANSCRIPTION {} | there are no stationary modes in a singularity but what there are are dynamical modes uh temporary modes if you will. |
TRANSCRIPTION {} | like half the country in the US can't put $500 together in an emergency. Okay, that itself is an emergency. |
TRANSCRIPTION {} | you've heard us talk about the idea that people end up doing four activities uh food, art, music and sex and the joke is not in that order |
TRANSCRIPTION {} | In 1937, the economist Ronald Coos uh asked why companies exist at all. He won the Nobel Prize for this. |
TRANSCRIPTION {} | His answer, transactions are expensive. Finding people, negotiating, and enforcing contracts. It's cheaper to hire employees and have them on the inside of your company rather than negotiating every task on the open market. Companies are... |
TRANSCRIPTION {} | MIT and Harvard researchers are finally catching up with Seem uh have asked what happens when AI agents make transactions nearly free... They're calling it the Cosian singularity. |
TRANSCRIPTION {} | in our 2023 book, um Exo 2.0, we we said then that the KOS's law was breaking. |
TRANSCRIPTION {} | take the mission critical function in Uber, which is to match driver and passenger. It doesn't happen inside the organizational boundary of Uber. It happens out in the wild. |
TRANSCRIPTION {} | AI doesn't just automate the firm. It attacks the economic reason for which firms exist and the shape of the firm... now essentially a firm becomes a protocol |
TRANSCRIPTION {} | the organization dissolves from being a human hierarchical centered model to a totally different world and so this is a huge shift the biggest sense we've had since the industrial revolution |
TRANSCRIPTION {} | the whole point of of cosian economics is transaction costs determine the size of the firm... So if transaction costs are going to zero that agitates in favor of the size of the firm going to zero. |
TRANSCRIPTION {} | Ted Shelton and I wrote this paper on what does the organization look like... and we coined a term called a fiduciary wedge. The reason you still need an entity is for liability for fiduciary for proprietary data ownership for learning l... |
TRANSCRIPTION {} | there's no reason why a company or firm or legal entity can't be owned by other agents which will eventually happen in Argentina and other places |
TRANSCRIPTION {} | wouldn't that agitate in favor of the exact opposite coian economics of firms growing larger and larger so everyone has access to those internal capabilities within the frontier labs. |
TRANSCRIPTION {} | you could already have a frontier lab running its own hedge fund strategy, and I'm sure they're doing that now that outperforms the market and just running that model. |
TRANSCRIPTION {} | Alex, you remember when Sam Alman said, "Yes, in the future I think an AI should be running open AI." |
TRANSCRIPTION {} | Sam, just in the past 24 hours also on social media expressed surprise, shocked, shocked that Open AI was able to solve uh uh Navier Stokes. |
TRANSCRIPTION {} | you end up with what Daria was talking about where one company like Anthropic will be all private enterprise right um and the that's unlikely to happen |
TRANSCRIPTION {} | it's really annoying when you go past 12 people and then past 150 people. This is where you kind of have that Carthaginian demon of disorder mock coming into organizations and they get misaligned. |
TRANSCRIPTION {} | I don't think you'll see the oneperson company. I think you'll see the 10 person company |
TRANSCRIPTION {} | economics fundamentally does need to flip from being scarcity based to being abundance-based because we can rearrange bits and soon atoms in any way that we want. |
TRANSCRIPTION {} | Meror now has what 50 to 100,000 individual actors that are effectively little companies in India and around Brazil now too. |
TRANSCRIPTION {} | Elon is building the single biggest integrated vertical company that the world has ever seen with complete supply chain control all the way down to raw sand turning into chips. |
TRANSCRIPTION {} | this week Tesla opened an official interest form for businesses who want to buy their own cyber cab fleets and build mobility hubs and charging infrastructure for the robo taxi network. Uh no pricing or delivery terms yet. |
TRANSCRIPTION {} | this can only work in my opinion because of the price tag on cyber cabs projected at $30,000, affordable to anyone who was previously an Uber driver. |
TRANSCRIPTION {} | we'll have a model which is kind of like uh some combination of Uber and Airbnb. So if you if you're a Tesla owner, uh you'll be able to add or subtract your car to the fleet. |
TRANSCRIPTION {} | decades ago would have been accumulating a laundromat or a restaurant franchise. Now it'll be yeah you own a fleet of robo taxis or a fleet of humanoid robots. |
TRANSCRIPTION {} | the central force wants you to succeed so badly to get the momentum going that they subsidize the hell out of your success. Like if you were the the fifth Starbucks owner, you would have been guaranteed success by the mother ship. |
TRANSCRIPTION {} | you think about Airbnb's marginal cost of adding a room is near zero. If you're high, you have to build a hotel. Same thing here. |
TRANSCRIPTION {} | It's the one with the lowest operating cost uh the lowest uh production cost and I haven't seen anything yet compete with uh with Tesla here |
TRANSCRIPTION {} | it's becoming increasingly clear robots are the biggest investment class that we'll ever see. |
TRANSCRIPTION {} | The global population over 65 is projected to grow from 852 million uh back in 2025 to 2 billion by60. That's more than half of all population growth over that period is coming from people over 65. |
TRANSCRIPTION {} | fewer working age people supporting more retirees is going to put huge economic pressure on pensions, on health care in every developed economy. |
TRANSCRIPTION {} | this is why I say longevity is not a luxury. It's an e economic policy for the century ahead. Right? An 80-year-old with the body and mind of a 50-year-old isn't a pension liability. They're a founder in this future economy. |
TRANSCRIPTION {} | the concept of education, career, retirement essentially evaporates, right? you have to have repeated cycles of learning uh creation and then taking a sabbatical |
TRANSCRIPTION {} | We've seen the preview with what's happening in in Japan and you you need it. China is moving to robots because they have to because of the one child policy and the population bomb that got coming towards them. |
TRANSCRIPTION {} | These numbers are hugely understated because they include India and parts of Africa that are still having babies like crazy. But if you look at China and you look at Europe, the numbers are much more acute than this. |
TRANSCRIPTION {} | America is largely immune because we have huge amounts of immigration right in the working age bracket. You don't tend to have a lot of immigrants that are 65 and over. |
TRANSCRIPTION {} | 150,000 people plus per day on Earth are dying. and putting an end to that, at least the beginnings of putting an end to that is what this looks like |
TRANSCRIPTION {} | the counterfactual looks more like the limits of growth from the Club of Rome, which I think was like early 1970s... based on the false premise that somehow humans are going to be overcrowding uh purportedly scarce surface area of the pl... |
TRANSCRIPTION {} | We should have more kids. Kids are wonderful and having a fully integrated care for kids throughout that can reduce the cost of child care and supporting and raising kids in the best way is probably the biggest thing anyone could do for ... |