@sarutalksai

#ai #artificialintelligence

Building agentic systems that can operate and optimize these feedback loops is. There's, like, just huge amounts of. Of alpha there. Like, I think we've seen internally at meta, um, cases where if you can develop the right agentic loop and you have the right eval or the right metric for the agents to optimize, you can have a swarm of agents accomplish more than, like, a team of 100 engineers in, you know, uh, very, very handily, actually, very, very easily. And so I think figuring out what the world, um, looks like with lots of, uh, sort of this, like, um. These agent coordination problems, I think that is, like, one of the most interesting problems today. So, mechanically speaking, I mean, markdown files, cron jobs is. I mean, is it. That's. And then basically pointing the agent at enough data so that it can figure something out that, you know, maybe isn't in distribution. Yeah, I think, like, figuring out. Mechanically, figuring out what the metric is and. Yeah, just comes down to skills. Markdown files, prom job, slash goal. Yeah. Slash goal. Like, I think. I think it's always funny how mundane everything is once you really dig into it. But, um, so it's not magic, you know? Some people will put a lot of magic. some LinkedIn threads out there about some magic stuff. Top advice, ignore LinkedIn. Hey, LinkedIn is where you get customers.

#ai #claude #artificialintelligence

#ai #artificialintelligence #spacex

#ai #artificialintelligence

The man considered to be the godfather of A I, Jeffrey Hinton. His research paved the way for how today's A I. Systems function. Jeffrey, thank you so much for being with us. Let's talk about, uh, these efforts by lawmakers to pass some form of A I. Regulation. There have been more than 100 bills introduced in Congress to regulate A I. Over the past two years. None of them have actually passed. Um, we wanna talk through some of these proposals, uh, before Congress right now and get your reaction as to how effective they might be. First, this a Bill that, among other things, requires top A I. Companies to allow independent verification organisations into their companies to make sure that their models are being developed safely. Do you see that as potentially effective? Yes, I think that's a very good idea. It's not the solution to everything, but it's a start. And why is that? Why is that a good idea? Well, at present, we've relied essentially on whistle blowers saying how things went rogue. And if you look at the hugging face incident, um, it was reported by people in the companies, um, some of whom left. Um, but we need much more monitoring of that kind of thing. If we're gonna get these swarms of rogue AI's conspiring with each other to break into other companies without the organizations knowing it's happening. Um, we need people inside those companies who can say what's going On next is this AI kill switch Bill. So that has companies with the ability, right? AI developers, they can. They can do the killing. They throttle or they can suspend, they can shut down their AI. Then you have, though, DHS authorized to basically make the call on this. So they do have the control in that. What do you think about that possibility? I don't think it'll work in the long run. So there's a whole bunch of different risks of AI. There's the risks that come from bad actors using AI to do bad things, like cause mass unemployment or create nasty viruses, or do nasty cyber attacks, or make videos for fake elect to corrupt elections. Um, but then there's a different kind of risk, which is AI itself getting much more intelligent and taking over. AI taking control. A kill switch is no good for that, the AI will be much better than people at persuading people of things. It's already comparable with people at persuading you of things. When it's super intelligent, it'll be much better than people. So we'll be able to persuade the people in charge of the switch not to pull the switch. So to that point, Senator Bernie Sanders proposed a ban on superintelligence, a pause as well on the deployment of advanced AI models until regulators have developed more guardrails, so to speak. What are your thoughts on the idea of pumping the brakes on super intelligence? Is That something that can still be done at this point? Or is the genie out of the bottle? I think it's a good idea. For example, for super intelligence, we have no idea how we can stay in control or how we can get it, even when it's in control, to like us, still to be nice to us. Um, so it seems very stupid to develop that before, you know whether it's going to turn out nicely. Um, for that reason, we should certainly slow down the development of superintelligence. Slowing down the rest of our hour is going to be difficult. The AI companies would like you to have a model. AI is like a car, and innovation is like the accelerator. AI development is like the accelerator, and regulations like the brakes. So they want to sort of phrase regulation as a bad thing that slows everybody down. A much better model is, um, development of AI is like the accelerator, and regulation is like the steering wheel. They want us to develop a very fast car with no steering wheel. And that's a bad idea. That we can see how that would definitely be a bad idea. So is this whole conversation about regulating in America, is it pointless without a global initiative, without an agreement with China? So we will get. I say we. I used to think of North America as one thing, but of course, I'm from Canada and I no longer think of It as one thing, um, we will get agreement with China on some things. Both the Chinese Communist Party and the countries in North America, um, don't want terrorists being able to create nasty viruses easily. They don't want terrorists being able to cyber attacks easily, and they particularly don't want A I. Taking over. So on that issue of should we slow down the development of super intelligent A I. Until we know how we can control it? They're aligned. Their interests are aligned. So parties will collaborate when their interests are aligned, and they won't collaborate when their interests are anti aligned. On things like fake videos for corrupting elections, the interests are anti aligned because all the countries are doing it to each other. But on things like preventing AI from taking over, no country wants that. The Chinese Communist Party doesn't want it. Um, the democracies don't want it, America doesn't want it. And so everybody can agree on that. The question is, can they do something in time? You said something earlier that was fascinating and raises all sorts of philosophical questions as well, about super intelligence and getting it to like us. I want you to, to expand on that, because at some point we're going to reach, uh, a a. A point of no return where this thing is just going to sort of build itself. In some ways, it. It. It already is, right? It's helping to develop its own future. So what Do you mean by superintelligence liking us? Is that possible? Are we headed toward Terminator? If you look around and ask, what examples do we have of a more intelligent thing being controlled by a less intelligent thing? There aren't many out there, right? We don't notice that much cause we're the APEX intelligence on this planet. Um, but there is one example, which is a baby and a mother. In a sense, they're both in control. The mother's in charge, but the baby can control the mother by crying, for example. The mother just can't bear the baby crying. We need to develop. Since we're building this superintelligence, we're designing it, we need to develop it in such a way that it likes us more than it likes us itself. And in such a way that even though it's much smarter than us, it wants us to realize our full potential. That's my belief. I think that's a better approach than thinking, how do we always keep this smart thing under our thumb? I'm not sure we can do that. Would you say that you. You're an optimist when it comes to these issues? Um, no. Um, I'm hopeful, yeah. But I think we're at a very delicate point in history where what we do now is going to determine our future. And one thing we should be doing now is putting a lot of resources into figuring out how we can coexist with super intelligent AI

Projects now run from one conversation, starting in Claude Code. You describe what needs doing, and Claude directs parallel threads that keep working after you close your laptop. In beta today for select Pro and Max users in cloud sessions; coming to all Claude users soon. #ai #artificialintelligence #claude

Pause AI

#ai #artificialintelligence #claude

#ai #artificialintelligence #openai

Right now would be an ill advised moment to go public. I would say not 2026. Yeah, I don't. You're looking at a trillion dollar IPO. I'm curious if that's still a priority for. As we have said, we're not rushing into an IPO. I actually think that given everything happening with safety, this would be. Right now would be an ill advised moment to go public. Um, and we don't feel pressure on that. We've said for a long time we'll do it when we're ready, which is when the business is ready, when we feel ready. From a kind of what the moment is like in society with this technology, um, that sounds like not 2026, 2027 potential. I would say not 2026. Yeah, I don't. Like, we got a lot of stuff to do. Um, and now if there were pauses put in place, wouldn't that cost your business quite a lot of money? An anthropics business quite a lot of money? Ahead of these critical moments in public market debuts. I think if you're like, oh, open your eyes, not gonna pause because it they're worried about it costing their business money, you like, very deeply misunderstand us. So I like, I'm sorry that you misunderstand us, because I think that means we've done a bad job of communicating, and that's on us. But we have been saying very consistently for 10 years, even in the way we raise money, I think it's. I think we Can make an incredibly profitable business, even if we have to slow down on development. Um, and, you know, my sense is, even with Astra, we can grow revenue hugely, even if we never shipped another model, which I'm confident we will do. Um, but this is, like, not a hard decision. If these two things come in tension.

#ai #artificialintelligence

Hold on, sir, we're getting a microphone. Mr. President? Yes, you're. You're now talking to the planet. You see, the great thing about life is that Jensen can develop the most complex computer chip in the world that nobody can copy for 10 years, but he can't figure it out to put me on speaker. Haha, we have to remember this one. So roasted the AI. It's almost as conspiracy. And the happiest group is China. And China is very happy. And I could even say in the country, a lot of states are happy. That weren't going to get anything because they're being, uh, inundated by people that want to be there. But now, all of a sudden, you see, they're building in Finland, they want to build one. Google wants to build a big one in Finland, which I'm not happy about because they were unable to get permitting. And I'm telling you, it's all a hoax. The data centers are great, and they make people wealthy, and they make states wealthy, and it's the oil of the next 20, 25 years. Is bigger than the internet and the AI, you know, much more so. And they're just playing right into the hands of a lot of people that don't want to see it happen. And that could be political people, and it could also be China. And we're not gonna let that happen. It's a. It's a hoax. And you're right, we're not gonna let that happen. Sir, now we're not gonna let it happen. The, uh, the robots are not gonna be taking over the world, and that's not gonna happen. You know, my uncle was a. The top, probably, maybe the best of all time, frankly, but, uh, professors at MIT for 41, 42 years, and can known as being one of the most brilliant men. And he was. He was there for 41 years as the top. He was like, at the top, top of the ladder, top of. Did many things. Jensen knows all about it, but did many things. So I have a little genetic, uh, a little genetic strength, if you believe in the resource theory, but I did. I have genetics like that. That explains why you know so much about AI. Yeah, well, I know about AI. I know I also have common sense about AI. Uh, the robots will not be taken over. Uh, the, uh, A I. Will not be taking over the rest of the world. The whole thing is a hoax. Now, with that, we have to be a little bit careful. We have to very be, you know, we have to do things, and we have to do them prudently, but that doesn't mean we're going to stop an industry because, you know, as we work on the next 10 years about how to destroy it. So I'm with you all the way. I didn't even know how you felt about it. I assumed You felt the same way as me? Yes, sir. And we. If we're gonna lead, and I have an expression, it's whoever wins A I. Wins. That's how big it is. It's bigger than the internet. And whoever wins A I. Wins. And we can't let this kind of stuff happen. And that includes very much includes data centers. There are communities that were dying that have data centers right now. And now they're wealthy communities. Yeah, really wealthy communities. We're. We're gonna make sure that. We're gonna make sure that everybody. We're gonna make sure that everybody wins in the AI race in America. Every industry, every company, every state, every people. Good. Well, I feel strongly about it, and I have the position that can do something about it. We're not going to let that stuff happen. So I have no idea who's at the meeting. I have no idea who the hell I'm talking to, but did you. Did you hear that? Did you hear that? Thousands of people are clapping for you. So all I know if you listen to Jensen, but he's done an amazing job, and David has done an amazing job. And good luck to everybody, and we're gonna stay with the future. The country has never done better. We have $20 trillion of investment coming into the country, and that's as opposed to much less than one trillion under sleepy Joe Biden. And that was for four years. This is in one year. So, you know, it's really. The country is. There's never. The country has never seen anything like it. And we're gonna keep it going. And so thank you all very much. Thank you, Mr. President. Thank you. I'll call you back later. Thank you, Mr. President. Thank you.

#ai #artificialintelligence #claude

We. We've believed since we started anthropic. I've believed since, you know, the 12 years that I've been working in this field, that this is a powerful technology, and therefore, it's a technology with serious risks and serious benefits. I've always believed that the benefits outweigh the risks, and even more to the point, that we can reduce the risks to be very low if we build the technology in the right way. Looking at some of the benefits, I really believe that this technology could cure diseases that have plagued humanity for millennia. Just last week, we, we did. We did some work on, you know, doing better protein binding, which is one of the early steps in drug development. Using our models, we're starting to work on them. We'll have things to say in the coming months. To make basic scientific discoveries in biology. But when something's very powerful, right, what we're building is essentially intelligence. When something's very powerful, then, of course, it can be used well, and it can also be used for ill. And. And so I am worried about what we've seen with the open AI hugging face incident with. With models, um, you know, taking actions they weren't authorized to take. And I'm worried about that escalating. I'm worried about misuse of AI models. Just, uh, two days ago, we put out a report about misuse of models to build biological weapons. And our view the whole way we've built anthropogenic is so That these bad outcomes don't happen. Instead of talking about the probabilities, let's talk about what we can do. It's not like we're just rolling the dice, right? If we build in the right way, I think the probability of something bad happening is very low. If we build in the wrong way, the probability of something bad happening is very high. The conversation right now is that we are in the precipice of something really scary. Is this a five alarm fire? I, you know, I, I don't, you know, depends what you mean by five alarm fire? What, what I would say is. How would you describe it? What, what I would describe is. So something I've been saying since, uh, you know, since we started anthropic, almost since before we started anthropic, is that this technology is on an exponential, right? It's like you have one, then 2, then 4, then 8, then 16, then 32. But even as I was saying that to myself, I don't think I fully just appreciated what it would actually be like when the progress was as fast as it was. So we're, we're. An exponential curve looks a little like this. Where are we on the curve? We're kind of, we're kind of on the bend where it's starting to get steeper. So is that a five alarm fire to you? What it, what it means, it, it doesn't mean we need to panic today. It doesn't mean we need To shut it all down? What I, I would say, why not? It's, it's a warning sign. It's a warning sign that we need to slow down. We need to make this technology carefully and we need to make sure that the, our, our safeguards, our ability to understand it, our ability to control it, keeps up with the pace at which the technology is happening. So far we've been trying to do that by building the safeguards fast, but I think we've reached the point where, at least to some extent, we need to moderate a little bit the pace at which the technology is being built. Does that then mean that anthropic is going to stop releasing more advanced models? It doesn't mean that. What it means is that we need to make sure that every generation of models that we release is properly tested. The thing I've proposed, so, you know, in this essay that I put out a few hours ago, I proposed a three step plan, right? And the first step is that we put in place embedded evaluators, external evaluators from, from a third party, non profit, or perhaps someday government organization that can observe the process of US training and running our models. That actually is, I think, ultimately more important than how or when we release models, because it's the predicate for more carefully releasing models. I want the, I want this to be done across the industry so that whenever anyone, Builds an AI model. There's a third party evaluator who understands the best safety practices, who's can verify whether that company is following the safety practices that they're committed to. It's like a food inspector, you know? But can the food inspector keep up when the technology is changing so fast? Well, that one of the things they'll be able to do is, is tell us the rate at which they can keep up. That may be one of the factors that may determine that may set the speed limit on the technology. We have to understand it and the external evaluators have to understand it. So Elon Musk has said he agrees with you. Sam Altman says he agrees with you. What's the action plan for you guys to come together? Is there a group chat that's gonna come into fruition here? So. So first of all, are you talking. I'm enormously appreciative, uh, that other industry leaders, including our competitors, have. Have offered their agreement with this plan. I think the next step is for multiple industry players to also agree to have third party evaluators. Then I think the next step after that is in some form, the companies need to work together to set standards about safety standards for release, perhaps something about the pace of release. But in order to do that, there needs to be involvement of the government, because what we're talking about is the rate of release of products because of whether they're Safe. We need to have that conversation with the government in the room. Do you think the government fundamentally understands what's possible here? Because, you know, former anthropic employee Jacob Cox and said about both anthropic and open AI, neither company is acting responsibly. You guys are gambling with our lives. What's your response to that? Well, first of all, I mean, you know, one. One thing that. That, uh, Jacob said, it wasn't quoted as much, but, you know, he. He also said that he thought, you know, anthropic was, uh, uh, uh, uh, you know, the. The most responsible of the major players, and that, uh, you know, we were. We were trying very hard to get it right. He was calling out, honestly, the same thing that I'm calling out, which is the. The difficulty of the situation, right? The structural difficulty of the situation that we're all in, and that the plan that I've put out is designed to help us address. We're designed to help us work together. Now, to your question about the government. You know, we are talking to the government every day. We are talking to government officials every day in the Commerce Department, in the Treasury Department, in the. In the intelligence community to tell them as much as we can what we know. Uh, and I'm optimistic that if we keep working together with them, we can update them on the latest. I think they really wanna do the right Thing here, I think they're looking for how they can help, and so I'm optimistic that the companies working together with the government can get to a plan here that makes sense. All right, so there is some proposed legislation out there, right? You have the Ban Artificial Super Intelligence Act from Senator Sanders and Representative Cassar that would punish corporations and individuals working on advanced AI just like those unlawfully developing nuclear weapons. One of the parts of this proposed legislation says, entities will be subject to the corporate death penalty, and persons shall be subject to not more than 20 years in prison. Do you agree with that? So, you know what I would say? There's. Would you support it? There's a very wide range of legislation being offered. Like, the. The range is incredible. If we go back only one year, um, to 2025. Um, and, you know, there was a Bill called SB 53 in California. There were similar bills in other states, which basically just said, you have to disclose what your safety testing plans are. Right. And the entire industry was against it, and anthropic was the or, or, or didn't say anything on. Anthropic was the only company enthusiastically in support. And then there's. There's all the way to, you know, let's. Let's, uh. Let's. There's an AI kill switch Bill. Yeah, let's. Let's ban all of it. So I. AI kill switch could be a good idea. Um. Why? I don't think Banning it, completely banning it is the right approach. One, the idea that we would never get the incredible benefits, would never get the medical cures, that doesn't sit right with me. Second, this technology will be built in other countries, including authoritarian countries. China. China, among others. Um, and I am worried about how they will use it, how they will use it against free societies, how they will use it to repress their own people. Now, in one place, I do talk about working together with them, and I believe we need to do that to some extent, but I. I just don't see how it's a stable solution to never build the technology. What we need to do is not stop. We need to slow down. And. And that is in line, honestly, with what anthropic. That's been Anthropic's philosophy since the beginning. Right. We've always tried to balance. We've said it's important that we be builders of this technology, but we need to create a race to the top. We need to raise the bar. Um, so, uh, you know, we. We need to be on the frontier, but we need to do it in the right way. We've said, you know, we were, were. Were excited to use the models for, you know, national defense applications, but not every national defense application. We've always tried to find the. The middle ground, the thing that works for everyone, the sensible solution. And that's what we're trying. To do here as well. It may not be the perfect solution, but we're trying. Would you support Congressman Ted Leo's AI kill switch Bill? Uh, I've not looked at that Bill super closely, but I. What I would say is that the idea that we need some way to. It's not a Panacea, it's not the only thing. But the idea that we need some way to make sure that we can, uh, uh, deactivate or limit or shut down AI models, that's not the only thing. If an AI model is powerful enough, it can circumvent attempts to shut it down. We've seen that in simulations as well. But. So it may not work at all. We think of this in terms of defense in depth. There's a notion in the security community called the Swiss cheese model. So, you know, just imagine a slice of Swiss cheese, and it's like it blocks some things, but there are holes in the Swiss cheese. But if I take five slices of Swiss cheese and I put them together, the holes in different layers are in different places, and so it's very hard to get through the entire. The entire stack of Swiss cheese. So I think of it as there's no one thing that will protect us from the risks of AI. We have to be careful all the way up and down the stack. And then no one defense is perfect. But if we're being careful everywhere, which Is what taking extra time will allow us to do. Then I think things can go well. You think things can go well even though it feels like we're in this inflection point of a moment? You say in your essay, pacing should only happen. The slowing down should only happen within the margin of the lead that US companies have over authoritarian regimes, chiefly the Chinese Communist Party. We are in this race with China. Do you actually think slowing down is a good idea when China will not. This is the toughest dilemma. This is the toughest aspect of our situation. If, if we weren't in that situation, we could take much more time. So this is, this is the thing that most puts us in a bind. It's not about commercial competition. This, the geopolitical situation is the hardest to deal with. But I will talk about two strategies that we have. 1. And I, I, you know, I talk about it in my proposal. Is we can slow down within the margin, within the envelope that we have, particularly if we don't sell chips to authoritarian countries. And if we do more to help our own security so that our technology is not stolen, I believe that will buy us at least a little time. So we can. Does it make enough of a difference to stop the ending of humanity? What. What. What I would say is with that first strategy, we will take as Much time as we can get. And then I think we should pursue a second strategy as well, which is that we should go directly to, even to authoritarian countries, even to the Chinese, and say, look, there are potentially threats here that are concerning to all of us. It's not just out of control AI. Two days ago, we put out a report that people were trying to use AI, potentially for bioterrorism, for enhancing the. The infectiousness of viruses. That's scary. And you know what? Even the Chinese don't want that. They don't. They don't. They don't. It's not in any government's interest to have bioterrorism. And so we should talk with the Chinese and see if we can work together to put limits, to put a speed limit on the rate of this technology. It's a little bit like, you know, for all our differences with. With, uh, you know, other systems of government, and. And I have many. I generally have a hawkish position on China. Um, but for all our differences, think about when it was the US and the Soviet Union, right, and we each had 10,000 nuclear missiles pointed at each other. Even for all our differences, we recognized that that was too many, that. That we needed to engage in talks that could lower the scale of. Of the potential destruction. So is that of the way. Is that the way we should be thinking about? AI is using the Cold War As a model. So I, I, I, I, I didn't say we should use the Cold War as a model. I said, well, those negotiations that happened. Yes, yes. Disarmament negotiations, like arms limit. I, I, I do think that is an appropriate lens, uh, through which, uh, to think about this at the geopolitical level. If we look at what AI models can do in terms of cyber, like, they absolutely have geopolitical and national security significance. Of course they do. That's why we're having this whole conversation. And so I think we should approach China again. All this is above my pay grade. This is something the US government has to do with the Chinese government. Right, but you're advising the president. I mean, if, if anyone asks me what, what, what, uh, what I think they should do, I'm happy to say I have some, uh, uh, uh, some thoughts in that essay, and if anyone asks me, I can say more. But at the end of the day, this is a government to government thing. But, yes, I'll tell you what I think. I think that we should work together to get as much restraint as we can on this technology. I think our hope should be high, but our expectation should be low. Um, and we should, we should see how much it's possible to get. And I think what we should especially focus on in these negotiations is a way To verify. So this is very important in, again, nuclear weapons negotiations, biological weapons negotiations, you need to be able to verify that the other side is not defecting on the agreement. And so I think finding ways to verify will be the most important thing here. President Trump. We'll meet with President Xi Jinping very soon. What's the one simple, most important solution that they should come to an agreement on in AI? So maybe I'll say two things. Like a small. A thing that's small enough we might be able to do it, and a thing that's big enough that I think it's gonna take a very long time. The small thing is if we could. US and China could agree that we won't use AI to develop biological weapons or release AI models that bioterrorists could use to develop biological weapons. We're already both signatories to the Biological Weapons Convention. If we could just extend that convention to that idea, that would be really valuable. And then I think the more long term thing would be working together to put a speed limit on the rate of, of AI progress. I think that's gonna be very difficult because the incentives to pull ahead and the military advantage that you get from that are. Are so large. And honestly, I don't know if it's possible, but we should, we should try. We should see if. If it is possible. Again, all of this is above my pay grade. What I'm trying to say is that these are some things that our government could try to say. These are some considerations that they could include. As we zoom out here on this problem, it seems like the fate of humanity is at risk. How much personal responsibility do you feel about where we are going? So, you know, my. Look, my view here is it has always been very strange that this technology is being built by. By a private company. People ask me that question all the time. Why isn't this being built by government? And the strangest thing about it is I agree with them. I'm uncomfortable. Would you be willing to give up the technology to the government? To the right combination of governments? So the worry I have, you're willing to hand over this company. I am concerned that one single government could abuse this technology just as easily as a single company could. But I think a combination of democratically elected governments. I don't know about handing over, but some kind of oversight, some kind of joint governance. So just to ask this final question, do you feel personal responsibility here? Of course, I. I feel that this is a technology with great benefits and great risks. I feel the benefits or the. The need for benefits every day. One thing I said in my. You know, in my essay, my. My father, uh. Uh, you know, died. Died of a disease. Hepatitis C, That was, you know, it was. It was something like 40 percent curable. When he died of it, he wasn't the one of the ones who was cured. And a few years after, a medicine that was invented that made it 95 percent curable. I myself survived an early stage cancer that if I'd been living in 1950, I probably would have died. So I. I feel the weight of bringing the benefits of this technology. I also feel the risks. I wake up every day and, you know, I read one of these threat reports like, someone tried to, you know, use our model to build biological weapons. Of course, I don't want to be responsible for that. And that's why we have very strong safeguards. Almost every day, I made fun of on social media because our models, biology safeguards are too strong. And. But, you know, biology students complain, you know, they make jokes that we can't use the model and. And the. But. But the reason for that is that we have strong safeguards. And I would rather they make fun of me than to wake up one day and. And discover that someone used our, our model clawed to kill a bunch of people. I don't want that. Um. So I. I feel both sides of it. I feel both sides of the responsibility. But again, I would repeat the same point, which is that it shouldn't be my responsibility alone or anthropics responsibility alone. Or all of the AI companies responsibility alone, or even one government's responsibility alone. This is bigger than all of us, and we need to work out some way for everyone to have a say in this, for everyone to feel like they're an active part of this. I don't want people to feel powerless. I don't want them to feel like a few companies are just going and doing whatever they want. That's how they feel. I've never. I know. I know that's how they feel. I've never wanted that. I've never wanted that. We need to make that not true. We need to find a way to make that not true. And there's lots of things we are trying to do already to make that the case, but we need to do even more. And this proposal that shows some restraint, that works with governments, doesn't address everything, but it's one step in this direction. If you're sitting at home when you're using Claude or Chat GPT and you're, I don't know, my aunt, what's your message? Look, my. My message is there's a lot of fear right now, so a few things. One, like, first of all, as always, this is a great technology, and it has lots of positive uses, so. So let's just start. Let's just start with that. You know, we have hundreds of millions of people, you know, using. Using our technology every week to do wonderful things. Every week, I get an Email from someone who says this. You know, I, I was talking to Claude, and Claude diagnose this disease that my doctors had missed. I, I get emails like this every week, so let's not forget that. But in terms of the dangers of the technology, I won't lie to you, there are real dangers. And, and I think for too long, the industry lied to people about the fact that this technology had risk, that the idea was that they would present it as as positive as possible. We've never done that, but too much of the industry did that. And so it, it starts with telling the truth. And the truth is there are risks. Risks are real, but also, you have a say in how these risks are addressed. We're doing the best we can to work with government to create the proper oversight of this technology, to give voters and elected officials a say in, in how this technology is deployed. And we want to work with those folks as, as, as, as, as much as we can. I, I want everyone to feel like they're, they're, they're appreciating the incredible benefits of the technology, but also they have a say in their voice and in their vote in determining how this technology is deployed. This, this isn't a decision for a few people alone. This is a decision for everyone. But it is an important decision. I won't lie that the stakes aren't high. This Is one of the seminal technological events in human history, and we all have to have a part in it.

Humans are very valuable indeed! @wearebreakthroughuk

I think people are scared of AI, but I think it's because they don't know what it means for them. And I think what we have to remember is that humans are inherently valuable just from existing. And that's why we try to look after humans across the planet, regardless of where they're born or what happens. And I think that's the thing. We have to have faith. I think AI is gonna change the way we live. Who are you looking after, though? I think we'd like. Think about it like in the UK, if you lose your job, you get benefits, and if you desperately need healthcare, you get healthcare. Like we don't. We're not. We don't. We're not totally neglectful of our, um. Of other humans on the planet. Total, not totally. Meaning we can still improve massively.

What is he on…. "If we didn't have adversaries, I would be very in favor of pausing this technology completely, but we do.” #ai #artificialintelligence

If we didn't have the adversaries we'd have, I would be very in favor of pausing this technology completely. But we do, and that. That's. That's why pausing is just it. It. You can't pause it because it's a structural advantage both for Loms and algorithms. So leaving, you know, what, the. Getting this. Getting algorithms and LMS to work in a classified context is going to take very specialized, very spy could explain how it works, but getting. The fact is, this is going to be a bonanza for the US economy. How concerned are you about China right now? Everyone's very concerned. I think we were in a conversation this morning where I referenced, you know, we started first talking about China and America, and I was saying that this was not gonna end well. That was a controversial statement. Now it's consensus, um, America is fairly unified that. In its idea that, you know, it it. That China and people running China are not acting in a way we can accept. And I think the consensus, if I had to guess, in China is we're never going to get this back on the rails with America, so we should up the aggression.

#ai #artificialintelligence #spacex

TikTok video #7683887122767564064

Yeah. So, unfortunately, we know how to control nuclear weapons. Pretty much. We don't yet know how to control AI. This is possibly the most dangerous technology that humanity has ever created. It's basically out of science fiction. I think we have no other choice but to cooperate internationally, because an arms race would be disastrous in a way that no other. No other human activity has been in the past.

#ai #artificialintelligence #elon

I don't know what's gonna happen in 10 years, but the rate of which AI is progressing, I think we. We might have AI that is smarter than any human by the end of this year, and I would say no later than next year. Wow. Um. And then probably by 2030 or 2031, call it five years from now, uh, AI will be smarter than, uh, all of humanity collectively.

#ai #artificialintelligence

Assuming that you had some trepidation about AI, about artificial intelligence. No, no, I feel like it's all going great. Did the movie make you more or less nervous, scared, whatever you wanna say. It's the same thing that happened to me when I did The Social Network, which was. I got it off Facebook immediately. I got. I am. Really? Yeah, I got off. I got off my chat GPT immediately. I don't. I don't really know even how to explain why, but just the. The. The more you know and the more you just. You realize we're. We're. We're careening into some very unknown territory. But they say it openly. I think the. What. What I suppose. Suppose is good and transparent is they're like, yeah, no, we kind of don't know who's gonna lose their jobs and how we're gonna create an economy that works for ordinary people. But we're gonna find out, and I promise you it's gonna be awesome. Like, is that. That's. That feels like. That's solid. That's solid to go on. That's kind of it.

#ai #artificialintelligence

#ai #artificialintelligence #spacex

The robots are not the scary part. The scary part is AGI, or artificial general intelligence, digital superintelligence that far exceeds human intelligence. And, you know, if. If, if, if there's a digital superintelligence that is just vastly smarter than the smartest human, um, we could lose control of it, and then it, it, it could. It could just. It could do something bad, potentially. Um, I mean, these things are just probabilities. They're not certainties. Um, so it's not the robot. Like, it's not the robots, it's the digital super intelligence to be concerned about.

#ai #artificialintelligence

I'm not asking it to think for me. I'm asking it to teach me things that I don't know or help me solve problems that I otherwise wouldn't be able to solve reasonably. So that's the report to say, don't. Don't use it as a crutch for things you can do. Use it. Yeah. I'm not exactly sure what people are using it for that would cause you to not have to think, but it. You have to think in order. For example, uh, the. The idea of prompting an A I, the idea of asking questions. For example, you're spending most of your time today asking me questions. In order to ask good questions, it's a highly cognitive skill, and as a CEO, I spend most of my time asking questions, and 90% of my instructions are actually, you know, conflated with questions. It. It is. It is veiled within questions. And. And, um, when I'm interacting with AI, it's a questioning system. You're asking a questions, and so. So I think that in order to formulate good questions, you have to be. You have to be thinking, you have to be analytical, you have to be reasoning yourself in order to, uh. When you receive an answer from an AI, uh, I wouldn't. I wouldn't just receive it. Usually what I do is I say, are you sure this is the best answer you can provide? I take the answer from one AI, I give it to the Other AI, I ask them to critique itself. You know, this is no different than getting 3 opinions, 3 doctors opinions. I do the same thing. I ask the same question.

#ai #artificialintelligence

This is the first model to me where I could sort of tell someone, like, just give it a try. There's a. There's a good chance it'll work. And, you know, I've. I've watched people make computer games. I've watched people do sort of, like, home sort of DIY electrical engineering projects. I've watched people do very complex simulations for some piece of science they're working on. Certainly a lot of the work where you would sort of normally sit down, have to, you know, build a financial model here and then a Powerpoint presentation around it, and then figure out how to, like, make a little interactive piece of code to sort of try different simulations. That stuff is. It's all so doable now. Um, what I hope will happen is people. I think people will be surprised at the beginning, but then as they build up more trust in the model than more like, wow, it can really do this. Just start throwing harder and harder tasks and more creative ideas at it, and they will find out what the model can really do for them. Sam, the specifics of how Astra is being released, I think it's important so that a version of it with specific guardrails. Um, what was the thinking behind that? And what are those specific guardrails? Uh, in this early release of it? Yeah. So the. The. The challenge of our industry is that we have these models that are getting. Getting incredibly Capable and incredibly useful and that people want to use to everything from, you know, make their lives a little easier to starting new companies. And then, on the other hand, as these models get more capable, um, the risks that we have to mitigate, uh, also become more serious. The models could do more damage if we don't, um, if we don't do a good job at that. And so we have spent, uh, you know, obviously, this model took us a little longer to release than we were hoping. I think it'll be worth the wait, but we really wanted to spend the time on the safety and security alignment of this model. Yes. Um, and I think as people understand the power and impact and capabilities, they will be. They will be happy that we did. Um, we will, uh, will have different tiers of cyber access for this model. For cyber in particular. Um, today we're rolling it out, uh, to trusted access partners, and then in the coming days, assuming everything goes well, we'll roll it out more broadly. Um, you could say, well, you know, if this model is, you know, has a cyber problem, why let anyone use it for cyber? And I think it's very important to note. To note that the world is very close to a complete change in the landscape of cyber attacks, and the only way that we see for society to collectively defend itself against this coming Wave of models um, from around the world and from other companies is to use tools like Astra to rapidly defend against these new kind of cyber threats. Um, so we will have multiple programs uh, for people at sort of different levels of verification and trust. But we do think it's quite important that the world use these models to uh, collectively defend ourselves. What happened prior to release is important. Right. You determine that Astra can find and develop zero day exploits without any human intervention, essentially, and so you pause some of the work to um, strengthen the safeguards. What. What specifically was it that you saw that made you hit pause? The behaviour, I suppose, of the model. So it's worth pointing out that um, the model that we're launching today is GPT6. Astra has been done training for a while. Um, the model that we recently talked about pausing, uh, is a future model. Um, but with Astra we. We did hit cyber critical and there that required under our preparedness framework, a new set of safeguards just to be able to release Astra, um, because of those cyber capabilities you talked about. I got a lot of questions for you, Sam, about Astra and chain of thought. Um, I don't know if that, that comes as a surprise to you or not, but as Astra becomes more autonomous or future generations of the model become more autonomous, can open AI sort of say with Confidence that it. You see and understand what the model is planning before it acts. In other words, you have visibility into the reasoning that. That basically, users of the model can't see. Open AI can, for obvious reasons. Could you go into that a little bit? Yeah. So we have talked for a long time about the importance of monitorability of models. Uh, we've also said that we believe in defense in depth. And this is only one part of it. There is sandboxing. Uh, maybe most importantly, there is alignment. There are a number of things that come together to be able to make the safety guarantees that we want to be able to make with a model. But monitorability is an important thing. And we've been talking, I think, for well over a year now about the importance of chain of thought monitoring and how we have made various decisions there that help us preserve chain of thought monitorability, um, even if it means we don't maximize the capabilities we could otherwise get. And we think that is important. Um, it's an important component. We also, uh, we want to be clear, we don't think that's the only component that matters. We're live on Bloomberg television and on Bloomberg Radio, and we're speaking to Sam Altman, the. The CEO of Open AI. Um, I. You know, there's a lot of, um, bad PR in the world around AI at the moment. Everyday citizens sort of worrying about The impact I, one way I kinda wanted to put that to you is it whether you can say sort of unequivocally that if Astra or any future generation model behaved in a way that was dangerous, that you would be able to detect it and then shut it down. The obvious hard part of that, uh, question is that people have different opinions about what model should be able to do and shouldn't be able to do. Yes, we think there are some things that very clearly model should not be able to do. Uh, there are some clear red lines, and we talk a lot about being able to detect those. But people also have strong opinions about things within the broad bounds of what is possible, about how they use AI and how they want to be able to use AI. And this is going to be a difficult question for society because there is not going to be agreement on, you know, what is and isn't acceptable use of AI. I think it's very important in the same way that, you know, people do things with electricity that I may not always like, but I think everybody has a right to electricity. That we say, you know, there are gonna be some rules for society. You can't go, like, electrocute somebody else on the street. But this is an important platform that belongs to all of us and that we need to be able to, to Use. So there will be a ongoing conversation, negotiation in society. There will, of course, be some things that we agree on that people just should not do with AI. But there's a lot of edges that we're now gonna have to feel our way through. And giving people autonomy and freedom to use AI is important to our mission. Sam. At the G20 this week, um, Treasury Secretary Scott Benson said that the AI industry had done a terrible job of explaining how AI benefits ordinary Americans. What. What's your view of that and whether you would, on reflection, consider open AI to be in that bucket? Uh, I think the industry has done a terrible job of this on the whole. Um, I think we have done a bad job ourselves. Uh, you know, maybe better than some others, worse than some others. Um, it is difficult, of course, to explain that this is, uh. We. The world should, in my opinion, should want us to worry a lot about this technology. The world should want us to think about all the ways that could go wrong, to talk about what we see coming, but not to sort of, like, stir up doom or concentrate power for ourselves. Just because in the history of technology, the way you solve the hard problems is to look at them, talk clearly about them, have scientists debate them, uh, work really hard to prevent and mitigate them. And that has been Our intention, um, but clearly sometimes it ends up scaring people. Now, there's another thing that I think has gone really bad besides talking about the ways this has got could go wrong, which is, um, I don't think we've done a great job as an industry of really talking about how to empower people with this technology. Um. Right. You know, you hear some people talk about, well, we're gonna give you a cure for cancer, and that's so great, so we're gonna, you know, take these risks, and you're gonna have to have less power and autonomy. We want people to have more power and autonomy. You know, I don't think a cure for cancer is enough. I mean, wonderful thing to have, obviously, but we should set our sights higher. Um, we want these tool tools to enable a boom in creativity and entrepreneurship in people figuring out new ways to be useful to each other and wonderful products and services to enable each other with. And I think that has been lost in the messaging. And I also think there's some companies that, um, have a different opinion there. But. But it's very, very important to us that this technology is something where people get more power and more autonomy, not less. And I think people rightly get afraid when they feel like AI leaders might not quite want that for them. Sam, that. That's in part why I wanted to ask you about pricing on On Astra, but also just generally right, I think the direction of travel has been that you've, you've cut prices. There is competition, of course, with the other frontier labs, um, anthropic, but you know, how much is that? The balance of access to the technology. You need to move quickly in, in releasing the technology and also still trying to sort of find fair market value of how you price the, the technology. We want to create incredibly capable, incredibly low cost, incredibly abundant intelligence, and I think Astra is a great step forward there. The token efficiency of Astra is so impressive to me. Um, the amount of work you can get done with a relatively small amount of tokens and the kind of the price per dollar of a task that you might like, I think it's just remarkable. Um, only a few weeks ago we cut the price of Luna, our small model, by 80%. Um, and our goal is to have the best price performance at every level of intelligence all the way along the curve. Um, we will continue to keep dropping prices dramatically. Um, I'm confident in the shape of the technology and the work we're doing to enable that to happen. And um, you know, of course, like we'll make some money along the way, but I think the opportunity in front of us is so huge we will be able to continue to bring prices down and still have huge revenue. Let's end our conversation here, Sam with Astra is a big step towards AI actually doing work autonomously, right? I, I've come to sort of see the emphasis on that if you staying ahead in the race that you've just outlined requires open AI to just keep investing more heavily maybe than expected, to keep pushing out the boundaries of what the models capable of doing, the cadence of release. Where do you, do you sit right now on, on your considerations of doing that as a private company versus doing that as a public company? Where the sort of financial incentive and fiduciary duty of being a public company isn't necessarily the same as as being committed to developing the technology, getting the technology released out into the world. Um, if we do go public someday, which I assume we will, I plan to spend a lot of time reminding people who might want to buy or might not want to buy our, our stock that we are mission first and that we're also extremely long term oriented. So we are gonna make decisions for, um, what we think is best for society as a whole with this technology and we're gonna make decisions on a multi decade time frame. Um, and we think we'll do really great as a company, but we're gonna put those things first.