[ { "i": 0, "speaker": "Speaker 1", "text": "It was almost exactly a year ago that you started at Meta leading the AI team." }, { "i": 1, "speaker": "Speaker 1", "text": "And I thought maybe you could set the scene for us just a little bit. Um I I'm wondering if you can let us know where you think Meta is as an AI company today versus where it was a year ago." }, { "i": 2, "speaker": "Speaker 1", "text": "Specifically, like what's the reputation of Meta as an AI company today versus a year ago when you got there?" }, { "i": 3, "speaker": "Speaker 2", "text": ">> Um yeah, I mean I think it's been a very exciting year for um for Meta and AI. I mean what I think when um you know, roughly a year ago uh a little more than a year ago Meta released um Llama 4, and while I was" }, { "i": 4, "speaker": "Speaker 2", "text": "still an exciting release, I think you know, um it wasn't quite on the trajectory that Meta needed um to be able to continue building a lot of the products and experiences that it seek to" }, { "i": 5, "speaker": "Speaker 2", "text": "build. And so, you know, we've uh been hard at work over the past year since I joined and since starting Meta Superintelligence Labs, you know, we've um been undergoing an entire process of" }, { "i": 6, "speaker": "Speaker 2", "text": "building a new scaling ladder for our models, uh developing new sets of infrastructure, uh and and new research to be able to power sort of a new series and family of models. And you know, back" }, { "i": 7, "speaker": "Speaker 2", "text": "in April, we released the very first uh fruits of that labor, you know, the new Spark models and an update to Meta AI." }, { "i": 8, "speaker": "Speaker 2", "text": "Um the reception to all of that was incredibly positive. It was even better than frankly we'd expected internally." }, { "i": 9, "speaker": "Speaker 2", "text": "Um we saw incredible gains in usage of Meta AI. It was you know, the top of many of the app stores and um you know, we're we're sort of working on our even larger models today, and uh and are" }, { "i": 10, "speaker": "Speaker 2", "text": "excited about what we'll be able to demonstrate to the world. We clearly, you know, are we're on a very exciting and fast trajectory, and we're excited to continue showing the world what we" }, { "i": 11, "speaker": "Speaker 2", "text": "produce. Um I think that uh you know, the AI industry has gotten very uh hot and competitive obviously in the past year as well, and you know, we take that very seriously, but we're really" }, { "i": 12, "speaker": "Speaker 1", "text": "excited." }, { "i": 13, "speaker": "Speaker 1", "text": ">> Yeah. Obviously, the goal you're spending in the same way that an open AI Anthropic others are spending. Do you feel like you guys are in that same tier? Is is meta, you know, is it open" }, { "i": 14, "speaker": "Speaker 1", "text": "AI Anthropic meta at this point or do you feel like there's still a bit of a gap cuz I I'd say certainly a year ago that was the perception." }, { "i": 15, "speaker": "Speaker 2", "text": ">> Yeah, we don't I mean, the new Spark model that we released is not at the tier of the leading frontier models." }, { "i": 16, "speaker": "Speaker 2", "text": "But we believe it's a very exciting data point on the trajectory and we expect the upcoming models we release to be quite competitive with the leading models in the world." }, { "i": 17, "speaker": "Speaker 1", "text": ">> Yeah, you called new Spark an appetizer an appetizer model." }, { "i": 18, "speaker": "Speaker 1", "text": "To to build that metaphor out, when does the entree model get here?" }, { "i": 19, "speaker": "Speaker 2", "text": "You know, will it be at that tier that that you guys ultimately want to be at?" }, { "i": 20, "speaker": "Speaker 2", "text": ">> Yeah, we are we are in process of cooking the >> [laughter] >> We're just going to the whole conversation is just going to be this this back and forth." }, { "i": 21, "speaker": "Speaker 2", "text": ">> Yeah, yeah, no, I mean, we're we're cooking it. We're excited to show it to the world when it's ready. We're seeing very exciting and promising results in the process of training it right now. So" }, { "i": 22, "speaker": "Speaker 2", "text": "we're quite excited about it and and I think there's a you know, overall we built the entire research effort around predictable scaling. So the so the entire belief of our overall research" }, { "i": 23, "speaker": "Speaker 2", "text": "effort was, you know, that in many ways the central belief behind the current modern AI boom is that as you scale these models, you will see get incredible results and get predictable levels of increased" }, { "i": 24, "speaker": "Speaker 2", "text": "capability. And so you know, new Spark it was an early data point on that scaling ladder for us. The next models we release will be an even greater point on that on the scaling curves and we are" }, { "i": 25, "speaker": "Speaker 1", "text": "really excited to show the world what we'll what we'll be able to produce." }, { "i": 26, "speaker": "Speaker 1", "text": ">> And what obviously you went from basically rebuilding the team, rebuilding the lab to this new Spark model in a very short amount of time." }, { "i": 27, "speaker": "Speaker 1", "text": "What is the biggest barrier from getting to that, the appetizer to the next level? Is it I assume it's not resources. You guys are spending a ton of money on this on this effort. Is it just simply time? Is" }, { "i": 28, "speaker": "Speaker 1", "text": "it talent? Like what's going to bring your models to that that frontier?" }, { "i": 29, "speaker": "Speaker 2", "text": ">> Yeah, it's you know, we we've talked about some of this in some of our our public blog posts and whatnot. But it's about continuing to scale the data, the compute going into the models, as well" }, { "i": 30, "speaker": "Speaker 2", "text": "as continuing scale with research. So continue to drive advances in you know, underlying you know, underlying research breakthroughs to continue driving forward the progress in the models" }, { "i": 31, "speaker": "Speaker 2", "text": "and building infrastructure to support all of this. You know, it it is this is in many ways a year where all the labs are dramatically scaling up their models and we are on we think of" }, { "i": 32, "speaker": "Speaker 2", "text": "much faster trajectory to do so because obviously we've been doing all this work over the course of the past year." }, { "i": 33, "speaker": "Speaker 2", "text": "Yeah, we need to build the infrastructure, we need to scale the data, scale the compute, train these large models and show them to the world." }, { "i": 34, "speaker": "Speaker 1", "text": "Yeah." }, { "i": 35, "speaker": "Speaker 1", "text": ">> I want to talk about model strategy a little bit. Before you got to meta, everything was open source. That was definitely the overarching strategy." }, { "i": 36, "speaker": "Speaker 1", "text": "Muse Spark model is not open source." }, { "i": 37, "speaker": "Speaker 1", "text": "I believe I heard you on a prior interview basically say that as you guys were testing it, it didn't feel safe to open source. Can you go deeper on sort of what you mean by that and how you made that decision?" }, { "i": 38, "speaker": "Speaker 2", "text": ">> Yeah, so one of the things that we did when as part of Meta Superintelligence Labs is we updated our what we call our advanced AI scaling framework, which is really our view of you know, what are" }, { "i": 39, "speaker": "Speaker 2", "text": "the risks that we see in developing these very powerful models and how do we want to handle those risks as we see them in early testing. And we published a lot of what we saw in the process of" }, { "i": 40, "speaker": "Speaker 2", "text": "training Muse Spark in our uh preparedness report. And some of the things that we saw is that it actually triggered some high-risk areas uh in the course of early training. Particularly" }, { "i": 41, "speaker": "Speaker 2", "text": "around bio risk. Um but also a number of the risks were were elevated. And um you know, this is something I think the entire industry has seen as the models have improved pretty dramatically over" }, { "i": 42, "speaker": "Speaker 2", "text": "the past year. So, we certainly aren't the only ones to see uh a host of these risks, you know, show up in in as we scaled up the models and as we sort of kept pushing the frontier of research." }, { "i": 43, "speaker": "Speaker 2", "text": "But um we saw these risks triggered and we realized, you know, I think fundamentally those um when we launch a model like New Spark in a product, we have a lot of ways to mitigate some of these risks and ensure" }, { "i": 44, "speaker": "Speaker 2", "text": "that we're able to launch it in a safe and responsible way. Um it's much harder to do that when you open source a model and um you know, uh people can use that model in all sorts of contexts that we" }, { "i": 45, "speaker": "Speaker 2", "text": "may not have full understanding of. So, we're in the process right now of developing um models that we believe are fit and safe to be open source while still maintaining as much of the" }, { "i": 46, "speaker": "Speaker 1", "text": "performance capabilities as possible." }, { "i": 47, "speaker": "Speaker 1", "text": ">> So, you will still do open source. It sounds like Llama though is not the brand or the the pillar that you're going to do. Is everything open source going to be Muse Spark or adjacent?" }, { "i": 48, "speaker": "Speaker 2", "text": ">> Uh you know, we uh we have uh exciting debates about branding internally and uh >> [laughter] >> nothing to share right now, but uh but yeah." }, { "i": 49, "speaker": "Speaker 1", "text": ">> Okay. Um the the big models that are coming that you've hinted at, um give us a general sense if you can, obviously, you know, each model, each company is is perhaps known for certain things. Do you feel as" }, { "i": 50, "speaker": "Speaker 1", "text": "though you're moving in a direction where Meta's models are going to be known for, you know, best in class at X versus Y? Like, what are you hoping to accomplish with with what you guys come" }, { "i": 51, "speaker": "Speaker 2", "text": "out with next?" }, { "i": 52, "speaker": "Speaker 2", "text": ">> Yeah, so so already in New Spark, some areas where we were really impressed by the capabilities, even those again like a a much smaller model than ultimately we we intend to train, were around" }, { "i": 53, "speaker": "Speaker 2", "text": "multi-modality capabilities. So, it's ability to handle images, video, audio, and that's obviously very important and critical for Meta's business. Also, its capabilities in health were really" }, { "i": 54, "speaker": "Speaker 2", "text": "impressive." }, { "i": 55, "speaker": "Speaker 2", "text": "And that that was very exciting for us, you know, health is an area that we view as really critical as we scale these models out to billions and millions of people all around the world." }, { "i": 56, "speaker": "Speaker 2", "text": "And then also a lot of the early results we saw in the ability of the model to create, you know, vibe code and create little games or artifacts or whatnot were very powerful. So, we are doubling down on some of these" }, { "i": 57, "speaker": "Speaker 2", "text": "and and continuing to invest into the agentic capabilities of the model. So, we're really excited for the upcoming models we release to be very very capable agents paired with a lot of these other" }, { "i": 58, "speaker": "Speaker 2", "text": "strengths around multi-modality, around health and many others. And ultimately, what we're really excited to build for the world is are the best personal agents for everybody around the world as much as" }, { "i": 59, "speaker": "Speaker 1", "text": "possible." }, { "i": 60, "speaker": "Speaker 1", "text": ">> I want to get to agents in just a second because you you just made some news on agents actually yesterday. You have other stuff in the works that we can talk about. But before we pivot off" }, { "i": 61, "speaker": "Speaker 1", "text": "safety real quick, I did want to ask a question just about China." }, { "i": 62, "speaker": "Speaker 1", "text": "You've been, you know, you've talked publicly about the risks and the threat of AI coming out of China. You guys have also trained on some Chinese open source models. I'm just wondering, can you give" }, { "i": 63, "speaker": "Speaker 1", "text": "us a sense of like how you view China right now in the in the, you know, through the lens of AI? Is it a threat in the way that we've heard historically? Do you feel like that's changed?" }, { "i": 64, "speaker": "Speaker 2", "text": ">> Yeah, I think that I think it is incredibly important for the United States to lead on technology and the economic benefits that can be created from AI. I think this is this is very very critical. If" }, { "i": 65, "speaker": "Speaker 2", "text": "you look at the history of civilization, you know, technological advances are very important for for countries or civilizations to adapt to and and be able to adopt and that really defines," }, { "i": 66, "speaker": "Speaker 2", "text": "you know, the really truly the course of history over over long arcs. So, I think it's very important that that the United States is able to lead on AI. Um and that's that's a huge part of our focus" }, { "i": 67, "speaker": "Speaker 2", "text": "at Meta as well as ensuring that we are able to contribute to the United States leading. Where do you think we are right now?" }, { "i": 68, "speaker": "Speaker 1", "text": ">> As a country, is the US leading?" }, { "i": 69, "speaker": "Speaker 2", "text": ">> Uh I think right now the US is leading." }, { "i": 70, "speaker": "Speaker 2", "text": "Yes." }, { "i": 71, "speaker": "Speaker 2", "text": ">> Okay." }, { "i": 72, "speaker": "Speaker 2", "text": ">> And I think it's you know, this is one of these situations where you you know, it's important for us always to track progress from many other countries, but especially China, you" }, { "i": 73, "speaker": "Speaker 2", "text": "know, be very thoughtful and understand exactly what's happening within each country and what are the reasons those things are happening, but I I think right now we're ahead." }, { "i": 74, "speaker": "Speaker 1", "text": ">> And what Sorry, I I said I was going to get the agents, but what one more." }, { "i": 75, "speaker": "Speaker 1", "text": "What could put that at risk? What could put that lead at risk? What's the most threatening um thing to stop that?" }, { "i": 76, "speaker": "Speaker 2", "text": ">> Um that's a good question. I mean, ultimately I think we are in a phase where um the the research advancements industry-wide that we're seeing from continuing to scale these models, apply more compute," }, { "i": 77, "speaker": "Speaker 2", "text": "apply more data to these models are just um incredibly uh exciting. And you know, in some ways the the progress and pace of research today is nearly miraculous. And so, um I think that it's important that we're" }, { "i": 78, "speaker": "Speaker 2", "text": "able to continue this pace of progress, that we're able to continue um you know, being able to uh continue scaling these models." }, { "i": 79, "speaker": "Speaker 1", "text": ">> Yeah. Um agents now, finally. Um you guys just yesterday, I believe it was, announced a business agent. Um so advertisers can use this to, you know, interact with customers. I presume" }, { "i": 80, "speaker": "Speaker 1", "text": "eventually help even um develop ad campaigns, things like that." }, { "i": 81, "speaker": "Speaker 1", "text": "But you're also developing a consumer agent." }, { "i": 82, "speaker": "Speaker 1", "text": "Talk me through your vision for how ultimately agents will reach all of us. Like I I guess I'm wondering am I Is an agent going to be similar to like my email address where I have one" }, { "i": 83, "speaker": "Speaker 1", "text": "core agent and maybe a secondary agent?" }, { "i": 84, "speaker": "Speaker 1", "text": "Or is it going to be like the apps on my phone where I have one agent for every single task in my life? Like what do you envision we're going to be using as a society?" }, { "i": 85, "speaker": "Speaker 2", "text": ">> Yeah, I you know, we ultimately think it'll probably land somewhere in between those. I think you know, we really believe that people are probably going to have one, maybe two, maybe a small handful of" }, { "i": 86, "speaker": "Speaker 2", "text": "of agents that they rely on. And maybe they have a personal agent that's focused on things like their health and maintaining their personal relationships and you know, helping them be a better" }, { "i": 87, "speaker": "Speaker 2", "text": "parent and be better with their friends and family. Um and then you know, perhaps they use that same agent in their work lives, especially if they're working in a small business or they're an entrepreneur or" }, { "i": 88, "speaker": "Speaker 2", "text": "um or um you know, uh you know, working within a smaller organization. And then you know, maybe there's worlds where if you work within a larger enterprise or a larger um company, then these become" }, { "i": 89, "speaker": "Speaker 2", "text": "bifurcated and separated. Not too much unlike email, let's say, or plenty of other uh you know, key technologies that we use from day-to-day basis. So um you know, we think that ultimately" }, { "i": 90, "speaker": "Speaker 2", "text": "uh it should be you know, agents will be something that are that become deeply personal um and should be things where over time you find yourself being able to rely on them more and more and more" }, { "i": 91, "speaker": "Speaker 2", "text": "for um more and more of your personal life, more and more of your work life." }, { "i": 92, "speaker": "Speaker 1", "text": "Um and that'll be a process that you know, all society goes through together." }, { "i": 93, "speaker": "Speaker 1", "text": ">> Do you feel that Meta in particular, and again, a lot of this happened before you got there, but long history of privacy-related issues, are people going to be willing to, you know, trust a meta agent with the" }, { "i": 94, "speaker": "Speaker 1", "text": "personal tasks of their life that that you're describing?" }, { "i": 95, "speaker": "Speaker 2", "text": ">> Um yeah, I think that this is like one of the most important societal question for agents writ large. I mean, I think that um there's incredible uh amount of innovation and technology that's we've" }, { "i": 96, "speaker": "Speaker 2", "text": "built out on things like agentic safety, um you know, uh ensuring that these agents are respectful to your privacy, ensuring that they're respectful of of your boundaries, um and continuing to design" }, { "i": 97, "speaker": "Speaker 2", "text": "products in a way that are able to support that. So, uh this is definitely something that we're taking very, very seriously and we're being quite thoughtful about. Um and ultimately, we're excited to, you" }, { "i": 98, "speaker": "Speaker 2", "text": "know, show the world what we've built, but um but yeah, we we think that this is not even just a meta problem. This is an industry-wide problem as we build more and more powerful agents. I think it it" }, { "i": 99, "speaker": "Speaker 2", "text": "is a redefinition of, you know, humans' relationship with technology in many ways, and that's something that um you know, we're all going to have to think through and work through together." }, { "i": 100, "speaker": "Speaker 2", "text": ">> How soon will we see a meta consumer agent?" }, { "i": 101, "speaker": "Speaker 2", "text": ">> Um we are actively cooking it." }, { "i": 102, "speaker": "Speaker 2", "text": ">> Cooking the entree?" }, { "i": 103, "speaker": "Speaker 2", "text": ">> Yeah, yes." }, { "i": 104, "speaker": "Speaker 2", "text": ">> And um Okay." }, { "i": 105, "speaker": "Speaker 2", "text": ">> But no, this is like I think this is one of the things that was very exciting for us internally about the new Spark launch in Meta AI that, you know, we even when those launched, we were cooking things" }, { "i": 106, "speaker": "Speaker 2", "text": "internally, both the larger models as well as um as well as some of these products that you referred to that we're if anything more excited about than what we came out with in in April." }, { "i": 107, "speaker": "Speaker 1", "text": ">> Okay. And how are you using agents right now? Your boss, Mark Zuckerberg, it's been reported he's has like a a Zuckbot essentially uh or various versions of of agents that he's tasking some of his CEO" }, { "i": 108, "speaker": "Speaker 1", "text": "duties to. Is there an Alexbot that's that's doing part of your job right now while we're on stage?" }, { "i": 109, "speaker": "Speaker 2", "text": ">> Um well, I definitely use agents to support and help me um in my work a lot." }, { "i": 110, "speaker": "Speaker 2", "text": "I think that in many ways, um you know, being a leader within a company is really about how well are you able to, you know, understand everything that's happening at the company at to the best you can" }, { "i": 111, "speaker": "Speaker 2", "text": "and help, you know, support your team and being able to continue to execute better and better. So, I think there's all sorts of things that agents can do to help you there, but I think the uses" }, { "i": 112, "speaker": "Speaker 2", "text": "of agents that are probably most exciting to me are the ones where I use are the ways I use them in my personal life. So, you know, I use an agent to help me be healthier. Um, and I use an" }, { "i": 113, "speaker": "Speaker 2", "text": "agent to uh, help me keep in touch with my friends and like ensure that I maintain those relationships. And I think these are use cases where they're like very notably different from" }, { "i": 114, "speaker": "Speaker 2", "text": "the world where I didn't use an agent at all. Like these are things that are I think are like hard have been hard historically for me to like stay on top of, you know, both my health as well as" }, { "i": 115, "speaker": "Speaker 2", "text": "um, keeping in touch with all my friends. And having an agent that's there to help support you do a good job with those things has been pretty transformational for >> Um, somehow we only have a couple" }, { "i": 116, "speaker": "Speaker 1", "text": "minutes left, but I wanted I want to ask some sort of bigger picture questions about AI and society." }, { "i": 117, "speaker": "Speaker 1", "text": "Um, but but I actually want to start at Meta, right? So, your team is getting an immense investment. Um, you guys are investing hundreds of billions of dollars in AI writ large. At the same" }, { "i": 118, "speaker": "Speaker 1", "text": "time, there were layoffs uh, at the end of last month. So, there are people and and some of the framing is, \"Hey, this is to offset that.\"" }, { "i": 119, "speaker": "Speaker 1", "text": "As the person in charge of AI, I'm just wondering how do you deal with that reality, right? That that you are working on a product that you're excited about, but at the same time, you know," }, { "i": 120, "speaker": "Speaker 1", "text": "the company is saying, \"Hey, this is costing jobs as well.\" Like how does that make you feel? How do you deal with that internally? I'm just wondering how you, you know, deal with that reality." }, { "i": 121, "speaker": "Speaker 2", "text": ">> It's incredibly difficult to say goodbye to teammates. And I think it's, you know, it's no, um, I think it's very well known that this is this is a challenging, um, thing to go through as a team and" }, { "i": 122, "speaker": "Speaker 2", "text": "something that, you know, is uh, is important to acknowledge. You know, we are ultimately really excited about the progress that we're making in AI and the products that we're building and, you" }, { "i": 123, "speaker": "Speaker 2", "text": "know, we're excited to bring those to the world." }, { "i": 124, "speaker": "Speaker 2", "text": "But, yeah, I think that there's, you know, running a company large company is very complex and, you know, we're working through a lot of those issues, but, you know, we don't take any of it" }, { "i": 125, "speaker": "Speaker 2", "text": "lightly, but, you know, at the end of the day, we're excited about what we're building." }, { "i": 126, "speaker": "Speaker 1", "text": ">> I think job loss in general is probably like one of the biggest fears with AI." }, { "i": 127, "speaker": "Speaker 1", "text": "What's your view on that? Is if AI goes the way that everyone envisions and we reach superintelligence, is there a world in which that can operate and live alongside all of us staying employed?" }, { "i": 128, "speaker": "Speaker 2", "text": ">> Um I mean, I think it's it's something that we should pay a lot of attention to and I think that we should, you know, track closely and try to understand what the impacts are. I think one thing though" }, { "i": 129, "speaker": "Speaker 2", "text": "that we never, you know, I think rarely talk about, but is also happening that's very exciting is that AI is enabling the creation of more businesses than ever before in the world and, you know, we" }, { "i": 130, "speaker": "Speaker 2", "text": "see this in our data. I'm sure many companies see this in their data. There are more new companies being started today be like through use of AI tools than ever before and that those numbers" }, { "i": 131, "speaker": "Speaker 2", "text": "are only growing and I expect as AI tools become more and more powerful, we'll see more entrepreneurs, more small businesses being started. And so, I think there's, you know, it's the" }, { "i": 132, "speaker": "Speaker 2", "text": "economy is a complex, you know, machine and one of the things that we see within our data today is that there are more small small businesses being started, there are more entrepreneurs, there's" }, { "i": 133, "speaker": "Speaker 2", "text": "more opportunity for entrepreneurs. And so, you know, and we're really excited about that." }, { "i": 134, "speaker": "Speaker 1", "text": "We're excited about supporting small businesses throughout the world." }, { "i": 135, "speaker": "Speaker 1", "text": ">> Is it the kind of thing that you feel we we saw the Trump executive order just this week where the administration wants to sort of review some of these models before they're they're released. I'm curious" }, { "i": 136, "speaker": "Speaker 1", "text": "what you thought of that, first of all." }, { "i": 137, "speaker": "Speaker 1", "text": "But two, is there another type of regulation that you think actually might be helpful to prevent, you know, job loss or prevent a world in which AI is is taking priority over the humans?" }, { "i": 138, "speaker": "Speaker 2", "text": ">> Yeah, I think well, I mean, first of all, I think it's it's this is a really important and powerful technology and so I think um it's it's great that the administration is you know, deeply considering uh what" }, { "i": 139, "speaker": "Speaker 2", "text": "how we should be thinking about the technology, how we should think about responsible deployment and and what all that entails. So, I think that's broadly speaking, I think it's been really great" }, { "i": 140, "speaker": "Speaker 2", "text": "that this has been an issue that the administration has been really involved on and very thoughtful of and >> Um you don't think it slows innovation to have a review process like that?" }, { "i": 141, "speaker": "Speaker 2", "text": ">> Um well, I think I think it's always a balance, you know, regulation is notoriously difficult to get right in almost all contexts, but um but I think that you know, what we what we're seeing from these" }, { "i": 142, "speaker": "Speaker 2", "text": "models is just that they're becoming dramatically more capable and I I think it's important that we're thoughtful about how we deploy them ultimately." }, { "i": 143, "speaker": "Speaker 2", "text": ">> Okay." } ]