Sequoia's Pat Grady and Sonya Huang declare AGI has arrived — and issue a call to arms for founders

Capital in the re waiting room. Welcome.

Welcome to the show.

How are you doing?

Cheers.

Good. Thanks for having us. It's been way too long, Pat. [laughter]

We missed you, too.

I missed you guys.

Most shows it's like, you know, maybe once a year we'll have someone on. I like that we can just get you every week. We'll start scheduling next week's appearance, too. Sonia, it's also been too long. We we enjoyed uh the candles that you sent us after our last uh hangout.

I'm so glad to hear the ideas for this year's swag. Just wait.

Yes. Is is the conference uh is still called a uh AI Ascent?

AI Ascent. It's It's gonna be on 420 this year. So that

Yeah, you were you were early to the whole ascend trend. Everyone's talking about ascending this year. And so I It's just a great brand for you guys. I love it. Um anyway, take me through the article 2026. This is AGI. Why were you writing it now? What was the one thing that you saw? What was the inciting element?

You want me to start?

Go for it. I I think there were two let me start with there were two purposes to the article and there was something that we saw that catalyzed the article. One purpose was to kind of put a stake in the ground

because there's this thing that people call the AI effect where once a technology is actually out there and working, we stop to believe that it's AI and we start to just call it by the specific name of the technology, you know, oh that's not that's not AI, that's just RHF, you know.

Yeah. And so, and so there's this um mythical quest for AGI and best we can tell we're there. Yeah.

And so we just kind of want to put a stake in the ground, say, "Hey, look, we're there." And and then you say, "Okay, well, what's the point of doing that? Why put a stake in the ground?" Well, that leads to purpose number two of the article, which is it is time. This is meant to be a call to arms for founders. It is time to take stock of what capabilities are available in the world today and apply those to real world problems as vigorously as possible.

And so there is enough great technology out there available to founders today to go solve so many real world problems. And particularly what we've seen over the last couple of months and this is kind of the triggering thing for the article um with cloud code and opus 45 and these long horizon agents which to us are kind of the tipping point that's gotten us into AGI like in particular what is possible now with that as sort of the third big inflection point of the last 5 years is pretty remarkable and so it's kind of meant to be a call to arms.

Yeah. We're not helping cuz we have physical goalposts here that should I mean we have to take the goalpost off if we've achieved

we no let's just mount let's mount it to the ground like we actually have [laughter] to

bring your your smell the AGI stuff was amazing

we'll put the goal

yeah see it's there we're there both congratulations

you're in the goalpost now you're physically in [laughter] the goalpost because you have declared AGI uh take me through a little bit of the history of SEOA a history of previous technological revolutions. I don't remember a uh a you know a 1999 this is internet uh but but as you reflect on the work the firm has done uh what what what can you tell us and what can you share about how the firm has processed these uh you know broad proclamations about the impact of technologies that are coming and then arrival dates when they are ready and that maybe changes the stance of the firm. Yeah, I I love that as a question. So, thank you for asking that. Um, we've been pretty quiet historically as a partnership and so this idea of sharing our thoughts with the world is is kind of new and hopefully it's a useful service to the founder community. If you go back to our founding, so 1972 when Don Valentine started Sequo Capital, he was the goto market guy at a company called Fairchild Semiconductor, which is a household name for people who studied the history of Silicon Valley, but but not really outside of that. It was kind of like the OpenAI of its day or it was the Nvidia of its day. Like it was the IT company of its day because it was the first company to um to really commercialize Siliconbased transistors. Like it was kind of the company that gave Silicon Valley its name. And as the go to market guy at that company, Don's job was to take this magical new capability and figure out how to apply it to real world problems.

Yeah. Make it useful.

Make it useful. And that sort of led him into venture capital because a lot of his customers were these founders and he just kind of fell in love with them in their quest to build these businesses. And so I mentioned that because we grew out of this semiconductor revolution where all of a sudden computation is a thing that was available to the masses. Um we move from that into systems in the next generation with the with the Apples and the Oracles um you know taking these chips and turning them into more computing systems. We move from that into the networking and PC era where all of a sudden you have endpoints on on every desk in every office thanks to the PC and you can connect them all up with networks thanks to Cisco and the router. We move from that into a more sophisticated form of networking which is known as the internet. Uh we move from that into applications which are a more sophisticated way to make use of the internet. Those applications start to show up on mobile devices and then eventually now that everybody has one of these mobile devices in their pocket there's an enormous amount of data. people started doing interesting stuff with data and now we have AI and so [clears throat] this lineage of Silicon Valley has kind of built up over the decades and each successive wave has built on the one that's come before and and being in this um very fortunate position to kind of be in the mix on a lot of these transitions and and actually being the only partnership that's kind of made it through these different tectonic shifts and kind of has this institutional knowledge you know gives us this broader view on Okay, what's the importance knowing what we know about past tectonic shifts? What's the importance of the current moment? And and with that, I'll hand it off to Sonia because Sony was the one who, you know, 3, four years ago prior to the release of Chat GPT kind of called this for what it was and put out the original generative AI post back in back in the fall of 2022.

Yeah. The other the other thing I think that's interesting about this moment is I feel like people have had this sense of like AGI will be here when it's just like so obvious in the economic data.

Uh and like that's like that's when it'll really feel like okay that's when you can declare it of like it's here it's working the economy and I think part of why I appreciated your guys' piece is it's like

it actually is going to require a lot of human agency and like human application of this. It's not just going to it it may not just like happen fully organically run running away by itself, right?

And full of arms, now's the time. Let's go.

Yeah.

Yeah. I wanted um I I I want to hear from you Sonia, but uh I wanted to ask like has there been any type of shift in in founders coming and pitching Sequoia with like an entirely

like I I I felt like last year I kept seeing like agent businesses that were exciting and the teams were talented,

but I felt like they were just in in many cases just like kind of rebranded SAS. It's like, okay, like I you're pitching me this as an agent, but like

if I log into the product, it's going to look like traditional software and maybe that's the this maybe maybe SAS is evergreen and we're just going to live in dashboards forever. But I've I've I kept wanting like wanting like more basically and I think we have started to see some new paradigms. But has there been any type of shift in terms of the companies coming in

where uh maybe somebody spent a few years at a lab and they're like okay we're actually going to kind of uh we've seen the future and now we're going to really reinvent the you know the entire product experience from the ground up based on this new set of capabilities.

I'll start but Pat is our app layer guy so you should you should uh you should talk as well. Um we we framed this at the bottom of our post like from talkers to doers and I think maybe that describes a little bit of what you're describing of like first iteration um geni companies um including some of the the great ones like it did feel like you know it was it was a it was the next generation of SAS was how it felt like um they weren't actually getting work done for you um part of the reason for this call to arms is like we're going from talkers to doers which is and I think it is both the long horizon agents is the thing but it's both the underlying model capabilities and the harness around it so like these models can now take action and iterate and persist in the world around us um and so what we're seeing now is that founders are selling companies that are not just doing the you know the you know newer brighter newer shinier version of SAS but actually something that takes real agency and takes real time horizons to execute. So, for example, a traversal in the troubleshooting space, like when there's a bug site goes down, data dog says like ding ding ding, there's an issue. Um, typically you have a war room that goes back and triages what the hell happened, goes back and scrolls through dashboards, goes through traces, etc. Eventually root causes it, fixes it. Um, this is a, you know, several hour to um, dayong process for companies.

It's funny to think of the it's it's funny to think of the human analog there. It's like if you have an employee that comes to you and they just tell you there's a problem and they're not [laughter] doing anything about it. They're just like, "We got a big problem here, boss." And you're like, "Okay,

okay." Like, "What are you doing about it?" And it's like, so traversal is like, "Okay, like we're actually gonna be what a good employee would do, which is like, we have a problem. Here's like how we're going to fix it, and hopefully this just solved before it even gets to you." Right.

Exactly. And so, in this case, there is no SAS equivalent to that. Right. This this it doesn't even feel like SAS. It feels like talking to a co-orker. And that's why the litmus test here is, you know, are you are you sell can you actually sell work? I'd say the founders that came into our door a few years ago, they all saw that vision. Like it's not it's not a new idea that you can build agents. I think what's new is that you can actually execute on that vision because these models are finally smart enough, capable enough to actually stay on the rails where you can just let them run.

And I would say founders have done a really nice job of making use of what is available technologically at the time. And I think I think over the last five years, we've kind of had three big inflection points. You know, inflection point one was chatbt when we saw the benefits of pre-training. And pre-training kind of gives you baseline world knowledge and some kind of instinctive judgment, so to speak. Inflection point number two was late 2024 when 01 came out and we had reasoning. And so the ability to kind of take your time and really think through particular questions and come up with deeper answers and better conclusions. And then inflection point number three has sort of been these last couple of months where we saw the impact of these long horizon agents. And so you've got some bas baseline knowledge. You can do some reasoning over that knowledge. Now you can actually bail, recover, stay on task, persist all the way through to some sort of an outcome. And I think if you map those three capabilities, the applications that people have built, it's no surprise that the initial applications were kind of like good Q&A or good summarization or good basic content generation. Like it was the stuff that pre-training sort of gives you. the next generation of those agents could could think a little bit harder. You know, like in the case of um uh in the case of notion as an example, which I think has done a really good job of staying on top of all this stuff, you know, you could now tell notion, okay, can you build me a database?

And it could actually reason through like what kind of database do you need? How should I build? And it can actually build you a database. It's pretty good. Um and then now with these long horizon agents, we're finally getting to the point where you can make a couple mistakes. you can take test a couple hypotheses and you can actually iterate your way toward the right outcome. And so look at um again you can look at notion and and you can effectively have your personal chief of staff doing stuff that you would otherwise be doing while you are in a meeting running in the background you know or or in the case of Harvey you can effectively deploy an AI associate after a data room or a deposition or some other legal workflow and actually complete it end to end while you're doing something else. And so I think this may maybe one way to put it is with this third inflection point, you don't have to babysit the model. You can just like give it instructions and let it go and you can have a few instances of it running in parallel. And you you kind of go from this is a cool tool that allows me to be more productive to holy cow, this is kind of like a team of co-workers that I have at my disposal.

Yeah,

totally.

Help me reconcile these two things that I'm feeling simultaneously. I agree with your uh declaration, the call to action, the just the idea of so much opportunity for founders to use AGI to create all sorts of different value and products and services. Uh at the same time towards the back half of 2025 I was seeing Ilia on Daresh talking about hey maybe we're in age of research and Richard Sutton and Andre Cararpathy talking about some of the more uh some of the more re restrictive aspects of the research progress and so is that something else that we're working on? How are you thinking about the progress that's happening in more purely research contexts? what even needs to happen there? Your role as venture capitalists in helping that happen. How do you blend those two things that both feel true? Um but are but feel somewhat at odds.

I think it's an end

and we're putting our money on both in both camps. Okay. Like on the research side, there's there's absolute fundamental research that needs to be done right now. Yeah. I'm personally I don't know if you've you guys have read Dave Silver and Rich Sutton's age of experience paper like that to me kind of outlines um the premise like we're going from you know learning from existing data internet data etc. Yeah, that was like the paradigm for pre-training to learning from experience, which is like reinforcement learning. The paradigm is you interact with environments, you get reward signals, um you improve.

Um and so I think there's a lot that's going to happen in terms of learning from experience and and um have been following you know what Rich Sutton's doing with Keen um and very excited about some of that direction. I'll give you an example also. We backed a company called Recursive Intelligence. Um they formally developed the chip at Google and this is like recursive chip design. So, similar to um in the game of go, you can have uh you can play, the agent can play itself and get superhuman at go, you can actually get superhuman at chip design and very surprising things come out of that. Like you end up having chip design that looks, you know, not like the Manhattan distance kind of like square grid stuff. You you actually have what looks kind of like alien like chip design uh layouts. And so, um we are absolutely backing research labs. Um but then if you like if you freeze research progress where it is I am convinced there is so much value to be built in the world and it's just like it's it's just lacking founders lacking agencies to actually go out and and and productize and and build companies around those like one of my examples. Go ahead. I was going to say uh how are you guys thinking about like more passive product experiences because people talk about agents and most of the time they're talking about like giving agents tasks like hey go do this thing and spend as much time as you need on it and uh you know we we'll see the result right like selling work but when you think about actually working with people some of the best employees are just without direct guidance constantly doing work in the background and coming to you only when maybe they need some type of like input. And so it's like more like you're just kind of reacting to the work that they're doing and and giving guidance. And so an example I would think of is

um uh you know with with Harvey it's like if somebody sends me legal docs

I should get a some type of notification from Harvey really when it's when it's already done some type of like

like I already jumped on it

diligence on it. I already did the work. You only so I'm not even worried about I'm just I'm getting a ping from Harvey. Hey, you should pay attention to this thing.

Yeah,

you're going to need to make a decision here. I've done all the work.

Um, and so and and even like with an email experience, like I would love to log into my email and not see just like the order at which the emails came in, but like here's three decisions you need to make. And again, like when you work with great people, they're going to if they have five minutes of your time, they don't just tell you here's the last 20 things that's happened. They're like here's the three things that you need to make a decision on. So, I'm really excited to see more of these like passive product experiences where the work is just being done and uh you only need to chime in when it's really necessary or if you're actually a bottleneck.

Yes, I I think this is a great question and a great direction and to me this kind of gets at the software as a service versus services as a software. you know that the nature of software is really changing as we go from an era of apps to an era of agents. In the era of apps, you want a lot of surface area with your customers. You want them to spend a lot of time in your product. You want to put workflows around them that are going to be really sticky that they're going to get accustomed to. In an air of agents, to your point, things can just be running passively in the background. The actual amount of surface area you have with your customers might be dimminimous. The amount of time they spend in your product might might be dimminimous. So the nature of the moat that you build is different. And the nature of the moat that you build, kind of what Sony was saying with the age of experience, it's all about context. It's all about like the environmental context of the job that you are trying to achieve and the feedback that that agent gets as it runs off and does something and comes back and you say actually, you know, like you did this part a little bit a little bit wrong. You need to do this part a little bit different. And I think in a perfect world you have something you have something like what Harvey does which you know Harvey kind of bridges that gap where there are workflows today for people who who want to do the work kind of how they've done it in the past just a lot better faster cheaper with the benefit of the AI brain and you can deploy agents to go off and do the work on your behalf and then come back to you when it is done and I think that I think that I think we'll see a lot of that over the next handful of years where there are companies who can kind of bridge that gap where they can live in the software world and have some of the workflows, better workflows because of AI, but some of the workflows people might be accustomed to and then separately they can deploy in the background with these passive agents that kind of function as co-workers who just come back to you when things are done. But it's two very different like design paradigms. It's two very different business paradigms. The fact that they're so different is one of the reasons why I think it's going to be tricky for a lot of the incumbent, you know, software companies to make the leap the same way it was tricky for a lot of on-prem companies to make the leap to cloud.

Yeah. Can you help me understand where you sit on the level of software eating the world? We talked to Tyler Cowan about one of the one of the problems with AGI showing up in economic statistics is that uh AGI needs to live in a digital realm. It has to affect the the the the digital economy, the services economy and that there's so much of the GDP is made up of the health care sector, nonprofit sector. He highlights a number of sectors that are sort of AGI resistant or AI resistant. And so that's why he has longer timelines for significant impact to the overall GDP figures. Do you agree with that or do you think that uh there's something different about AGI that should allow uh tech to finally move the needle on those stickier industries?

Well, I do think as like I I fully agree with the premise that you know this AGI stuff is only good it's only embodied in the digital world right now, right? And so like we we seem to be you know getting tantalizing close to solving the entire digital world. That is very exciting. At some point the physical world does become the next bottleneck. I think exciting thing is actually that like hey robotics is also going through a renaissance right now as well. Like some of the smartest people are going out and working on robots. And so I do think like similar just the pace of progress is so high right now. And I didn't think we were going to have digital embodiment that was reliable this early and now you can just let the thing run on your machine. Um, I think that increasingly we're going to get to physical embodiment and we're going to have a smooth curve there. Um, I I do agree with the pre premise of like if you want if you believe in the super fast takeoff recursive self-improving like you know everything is just going to you know go vertical line like I I just yeah we're not going to get there. We're not going to get there without the physical world. We're probably not going to get there even if we have the physical world. I just think the reality life is is really messy. Um, so I don't think we're at we're at ASI. We're not at that kind of like takeoff moment whatsoever. That's not the stake in the ground that we're putting. What we're almost saying here is, hey guys, this this quasi religious concept of AGI like is the thing, you know, generally intelligent and, you know, instead of just waiting for it to arrive, it's here. Let's let's let's build that. That's here.

Yeah. so much there's there's such a a there's kind of this toxic idea that uh I feel like is ingrained in so many people's head which is like AGI is just going to do that

like AGI is just going to do that

AGI is going to cure cancer

and it's like

it's at least in the near term from everything we're seeing it's going to be a human properly you know a group of of talented people properly using the tools to solve all the most pressing problems that we have and so it's and for young people

extremely paralyzing to be like in 2027 I'm either dead or a billionaire retired on a space yacht and so I could probably just chill for the next two years because it's one going to be one or the other complete abundance I worry that we'll have this sort of like lost generation of people that have let be ingrained in their head that uh it's not worth trying it's not worth

now is the time to build enterprise SAS even heard it from [laughter] Sequoia Capital get in the arena folks this is called action

I feel bad because I feel bad cuz a a friend of mine was pitching me this idea of like an an an agent to help consumers switch between uh like insurance products cuz like an insurance company will will sell you in a policy and then they'll just kind of jack up the price a little bit every year because they assume a lot of people aren't going to churn. And at the time I think chat GBT agent had come out and I was just like dude like like the the labs are going to like oneshot this problem like right like I think pretty quickly. I think you could work on this for

like you know six months and get steamrolled and like it's very possible that like if he had just focused on that one little problem you figured out something unique.

Yeah. And it's a big category and so I I felt bad.

There's lots of examples. How much do you guys um I wanted to ask like like processing signal cuz obviously there's so much noise right now and something that I've appreciated is

kind of comparing like the X timeline to the app store charts because

every single day on the X timeline like vibes are fluctuating and some news will break or some product will be released and the common response is like it's so over for XYZ company and then you look at the app store chart and it's like the company that just launched the thing is like not actually even in the picture at all. Like they're not in the top 25 of any category. And so like I feel like there's this insane disconnect between the real world. The real world is using chat GBT all the time, but they're clearly not paying attention to X, our little bubble. So it's like where are you guys really looking for signal and noise, especially in the later stages where you know the you're you're investing in companies that should be out in the real world at at some point. Yeah,

totally. Like before I get to that point, I I think it is so amazing that like all of this AI discourse is happening on on X right now. Like you could imagine a world where you know all this dialogue was like happening and behind closed doors or um you know scattered across different channels. The fact that it's all in X is amazing. Um and you know it's a combination of research and hot takes and what's happening in the labs. It's like it's I think it's awesome that it's all happening in one place. It's remarkable for young people that can just like jump in and learn by osmosis and understand who

Yeah, it'd be bad if it was only happening in Valinor. [laughter]

Totally learn. I still haven't figured out what the Elves comment was about. Um but anyways, putting that aside, like yes, like our job is to like, you know, to kind of try to pe signal from the noise. Um we have we're obviously doing everything we can in the background, credit card panels, uh web signals, etc. And I'll say like sometimes we see a you know trending web signal take on on X and it's just it's just wrong. Like we look at our our own sources of ground truth. The example like this some of the stuff on you know chat GPT is losing uh losing market share. We actually cut it by like US G uh US web traffic. It's a very different story. And so for us it's like it's really important that we're onX. I think all of the all the discourse happening here very important. But then like let's let's get to ground truth. Well, most importantly, just know that as venture capitalist, if someone refers to you as smog, it's sort of a compliment, but also sort of a dig. [laughter]

So, that's what you need to know in terms of Lord of the Rings references for for where you sit in the ecosystem specifically.

Is that like Smeaggel?

Smog is the dragon that hoards all the gold and is an antagonist in The Hobbit, but does have the gold, which is kind of like what you do. You know, you got a lot of capital over there in Sequoia Capital. Sorry, Jordy. Continue. No, what were you gonna say? Pat, sorry.

Oh, I was just going to say, you know, to the question on signal.

Um,

yeah,

you know, interestingly, one of the big themes over the last few years on on X, thanks to Lulu, is the go direct idea.

And I think for us, go direct is a little bit of try to, you know, let the world know what we're thinking, but probably even more so, just go direct to the founders. You know, the best source of signal we get is the founders. We spend, you know, we spend many hours every day just meeting with founders out there doing interesting things and we kind of have the ability to do primary research that not everybody gets a chance to do and I think that's that's still the best source of signal.

Yeah. What's uh what is changing in the AGI age in terms of adoption and uh just AI diffusing itself through the economy? This is something that obviously every company has to deal with because they can build a great product but then they need to get it into businesses if they're in enterprise SAS. They need to get adoption, make sure that it's being used correctly. There's the forward deployed model, there's self-s serve models. Is anything changing there structurally or is it all just uh you know the same playbooks? I

mean one thought is like I think we're going from over the last decade we went from salesled growth to productled growth.

Yeah.

I think we're going from product growth to agentled growth.

Okay. And I think you see this most clearly actually if you're using for example cloud code actively like it says hey you should use for database you should use superbase uh for hosting you should use versel etc and like it's choosing for you um the stuff you should be using and um I think increasingly like so it's it's most obvious to me with code like your agent is choosing your infrastructure um but I think it's going to happen across the entire economy like your chat chip is going to tell you like hey this is the um this is the place this is the travel you should be booking this is the you know and and so I think we are going towards

I think back to channel channel sales like old school

yeah old school but I think that the difference is like you don't have competing incentives here um and I think the open ad guys tried to lay like principles that I think are it's a very principled take on you know how do you want to tell us what users what to use and I think product growth brought us closer to the vision of best product wins but like ultimately people are still lazy they can't read all these reviews they kind of default to like what looks cool on the website. Whereas agent le growth like your agent has remember your agent has infinite time to go and make these choice for you and so it can go and you know read all the documentation read all the user comments like figure out for your use case

also has an experience like building on top of the infrastructure it's like there like eventually it's like well you want reputation reputation management is going to be very important the way you show up in these LLMs is important both in the pre-training and in all the research that they do and then also uh this whole idea of like, you know, SAS companies that don't have customers, they have hostages. Uh, well, if it's one click or one prompt to replplatform from the database that has that was trying to keep you hostage to the database that's trying to keep you happy.

Yeah. But it's it's also like I think it's good alignment. Compare it to like if you're building a home. Yeah. And like the the person like, you know, like running like the the construction firm that you choose like uses really bad cement and then you have like a terrible foundation. Like you're going to be mad. You're actually going to be mad at at like the the firm because you're like, "Why did you pick this? should have known this is like and so I do think there's a good alignment where it's like the best like you were saying the best product should win

because the the agent is incentivized to make sure that you're on the best like footing and using the best product and

cost is a factor uh quality reliability all these things so I do think there's that good alignment

that makes a lot of sense

the agent kind of accelerate this you know best product wins notion and best product for the human is maybe different best product for the agent sometimes

fantastic well thank you so much for taking the time to come and chat with us today. Uh, have a great weekend and we'll talk to you soon. Pat, we'll see you next week. Sonia, [laughter]

I can't wait.

Two for two for two.

Have a great to see you guys. Have a great Friday.

Goodbye. Phantom Cash. Fund your wallet without exchanges or middlemen and spend with the Phantom Card. Meet Phantom Cash, an

Andrew Reed special,

a Sequoia Capital Company. Um, we have some breaking news before we bring in Sean Frank. Uh, Sam Alman has taken to the timeline. E posts the melting smiley face emoji with a screenshot from Elon's court filing. The the in the court filing, here's what it said. Musk insisted that any new entity quote support the nonprofit mission and that open AI remain quote essentially a philanthropic endeavor. Now Sam Alman's putting it in the truth zone. He says, "Here are the actual September 2017 call notes from Elon. Uh, coming weeks top priority. Got to figure out how we transition from nonprofit to something which is essentially philanthropic endeavor and is BC Corp or CC Corp or something must tell the story and not lose moral high ground absolutely vital." So, we'll see if we'll see how Elon reacts to that. But if that is the case, uh Sam Alman is upset about it. He says Elon is cherrypicking things to make Greg look bad. But the full story is that Elon was pushing for a new structure and Greg and Ilas spent a lot of time trying to figure out if they could meet his demands. I remembered a lot of this, but here's a part I had forgotten. Quote, Elon said he wanted to accumulate $80 billion for a self- sustaining city on Mars and that he needed and deserved majority equity. He said that he needed full control since he'd been burnt by not having it in the past. And when we discussed succession, he surprised us by talking about his children controlling AGI. I appreciate people saying what they want and think it enables people to resolve things or not. This is Sam speaking. He's no longer quoting. But Elon saying he wants the above is important context for Greg trying to figure out what he wants. And Hope's revenge says, "Oh my god, girl, he's such a snake." [laughter] The drama continues. The drama continues on the timeline. Um, well, let me tell you about Apploving. Profitable advertising made easy with Axon.ai. Get access to over 1 billion daily active