Who Wins The AI Assistant Wars? + Why Didn’t Google Build Muse?

0:00 Who will win the great personal AI assistant war? And why didn't Google build its own version of Muse? That's coming up with MG Seagler right after this. Welcome to Big Technology Podcast. It's the first Monday of the month, and joining us for his recurring spot is MG Seagler of Spy Glass. We're going to talk today all about the personal AI assistant war which was brewing at the beginning of September but now has basically escalated into a full-blown battle war legendary tech face off that will continue to resonate for months after this. And we're going to also discuss the one glaring company that doesn't seem to have a competitive product here which is Google.

0:40 Although you could argue that some others are are there but Google especially should have it. So in we go. MG, great to see you again. Welcome back to the show. >> I am really hoping my computer doesn't crash during this recording because I'm running so many assistants right now. Like it's if you just look at my activity monitor on my Mac, it's it's it's near meltdown. I have a good amount of RAM, so I'm I'm okay, but it's I'm pushing some limits here. Now, this is not a sentence or not an opening you would have anticipated at the beginning of the month because or the beginning of last month because in September on the 1st of September you wrote this, no one has nailed the true personal assistant experience. And while you can finally at least squint to see elements of it, it's still hard to see how anyone will in full. Now, we have obviously Muse for Meta. We have Instinct, which I think you wrote a little bit about in the piece. we have open AI dots, we have Grockbot. I guess Microsoft has entered the battle in its own way. so personal assistants are the flavor of the month. And it seems like and actually be curious to hear your perspective on this that it's not just going to be a flash in the pan. It seems like this is a new product experience that we're just seeing the early days of that is going to stick given the advancements in AI capabilities. So, I actually would love to hear you at at the beginning of our conversation today, you know, go back to your thinking there on September 1st and talk a little bit about what's changed or if anything has changed and where you think this current moment in the AI personal assistant wars is going to go and then we'll get into who might win.

2:16 Yeah, that post sort of the impetus behind it was some somewhat driven by instinct as you mentioned because I felt you know it feels like months ago already but even just a few weeks ago certainly on all the social media feeds I feel like a lot of VCs in particular were were just touting instinct which is a new startup in the you know aentic and specifically sort of personal assistant use case role and so we saw a lot of a lot of love for that now you know this is a world I come from. I know how this operates like VCs feel like they got to give some love on social media so that they have any shot of of potentially getting their foot in the door if there happens to be a deal coming you know down the pike. And as >> this is a good secret of V VC Twitter.

3:03 Are you saying a lot of what we see on BC Twitter is just attempting to some around? >> I am I suggesting that some of these tweets are are have have ulterior motives behind them? you know, I can't imagine that that would ever be the case. you know, that some of this might be financially motivated perhaps. but yeah, I mean obviously you know you read these it's like oh my god is this the greatest thing ever? Instinct is cool. like and so I signed up for it, got access to it, you know, to their credit, they were ahead of this trend, ahead of you know, which we'll get to some of the other players like coming in.

3:39 >> can you describe what it's like? Like what the user interface is like, what it gets done for you. So, Instinct, I feel like, >> okay, so Instinct, first and foremost, again, going back, I feel like the impetus for this piece was basically I saw Instinct blowing up at least on the social feeds, and I feel like me with in my own head, you know, having been a reporter for all those years, been a VC for all those years, dating back a couple decades, I've seen this story play out a number of times. I just felt we were seeing some inklings of the same things that we've seen before.

4:12 and you've been in the same boat like you've seen a lot of this stuff before and so it felt like okay well what is actually going on here why is this time potentially different if it is or is it not different at all and so again so I signed up for instinct instinct I feel like first and foremost there there's sort of one of their obvious core breakthroughs is just the fact that all these models now exist right and so there have been you know long time ago years and years ago people have been trying to do sort of these personal assistant AI you know for everything from way back in the day on PCs before we had you know smartphones even and and you know there was that old famous concept video out of Apple for it was called na nav know knowledge navigator in the '90s if you haven't seen it Google it and and look it up on YouTube it's a great old school video of like this this assistant running on a futuristic Apple device which sort of you know is akin to an iPad and this face this you know this personification of an assistant, a virtual assistant. And you know, in some ways it is you know, analogous to what we have right now, which is interesting, but obviously it was way too early. So fast forward to, you know, what we're seeing with Instinct you know, out of the gates and it felt like again behind the scenes now we have the actual models in place being LLMs have risen to, you know, sort of take over AI at least for the time being. And so it feels like, you know, of course, everyone was going to try to go after what has largely been, you know, what I write about the holy grail of of AI, at least from a consumer perspective, fulfilling this personal assistant need, Instinct, I think, wisely recognized that they could sort of get off the ground running fast by simply leveraging already existing messaging paradigms. So rather than building, you know, a whole new app, whole new suite of of apps across all the different platforms, they basically just said, "Hey, use what you're using already. Sign up for our service and just start messaging us just like you would you know, either a friend or or potentially like a real life assistant and tell us what you need and and we'll work on our end. You don't have to worry about anything.

6:18 We'll work on on the back end to to to get done whatever you you message us with." And I think there was a lot of magic for people in that experience like not having to work through all these ideas of like you know hooking all these things up and figuring out like what works what doesn't and sort of just going and I think their other key insight in those early days at least was also being super proactive so you know and we can talk a bit about it why I think it hasn't worked previously but a lot of these things just have you know sort of like a blank page problem meets a boil the ocean problem. So, you know, the blank page, you don't know where to start with these things, right? And that's especially true, I feel like, in the age of LLMs because, as we know, you can ask them anything and they'll respond with something. That hasn't been the case historically, right? They've been very rigid in what they could respond to or not. dating, even with Alexa, right, in the last instantiation of of sort of some of the the voice-based AI tech, it was much more rigid and you could only do certain things. Now, you can ask them anything.

7:17 So, what what do you ask when you can sort of ask for anything? And then again like the boil the ocean problem which is long I think you know held back this space which is just they all try to do too much out of the gates like when you say again like yeah just just start and and see what you can do. Now I feel like you you get around that problem again with with some level of prompting. so basically the the assistant will come at you and say like hey I noticed that you know you have a message from soand so in your inbox would you like me to respond to them? as simple as that. but it can be a lot more in-depth than that, too.

7:52 It's like, hey, I noticed this bill is due, you know, in in 10 days. I noticed you haven't set up recurring payments for it. Would you like me to say log into that site and set them up for you? you know, if you're if you give me your credit card information and we can get into some of that. but again, that that was sort of the kickoff point. Instinct comes out and then lo and behold, we start this this wave of things that come after that.

8:15 >> Yeah. And so you know when you tried Instinct out in the begin in the beginning I think there was some skepticism that you had about what AI agents could do and you actually wrote about it. You said I keep coming back to the idea that this first killer use case or the first killer use cases may be more fun in nature letting agents do lightweight tasks for you that result in an absolute delight like opening up your to-do list and getting more comfortable over time with the technology taking over. assuming it can. and it will be baby steps, not one big handover. So, before we get into everybody else coming in, I'd just love to get your perspective on like what you've used it for and, yeah, what you think the potential is here and then we're going to go basically, you know, open the floodgates to the competition that came in starting with Metamuse.

9:02 >> Yeah. Again, that speaks to just that sort of blank page problem and the boil the ocean problem. they're both like they they sort of push against one another here in this space in particular. because I feel like people are willing to try something out. I mean, you know, obviously dating back to to websites and apps and now with the with AI, people are more than happy to try things out. but the the reality is like that very few people you know, if they don't get some sort of immediate payback on it, like they'll just move on to the next thing because again there are so many things right now. And so, with a product like this, you know, with what Instinct is offering, I feel like it needed to offer something to people right out of the gates, one of the issues with that, which I hit upon is that the reality here, again, because this is such a broad space where there's so much to be done. I'm not sure that the the killer use case is going to be or should be the same for everyone. Whereas before I may have said we may have even talked about it previously. I would have said like they should nail like you know getting movie tickets or something you know that's like a simple pretty defined use case. But here I do feel like that maybe instead they there should be a focus on say doing five to 10 things some of which will resonate with some user base some of which will resonate with others because I don't think it's going to be as simple as sort of just doing one thing really well. I do think you have to do a number of things really well. That said, I still feel like there needs to be the thing that you tell people, oh yeah, it did this thing for me and it's sort of the viral word of mouth mechanics and it gets people both trying out these new products, but also there there's a level of trust that needs that's needed for these products, right? And so if you know, you've heard from a friend that they're using this product and that they obviously inherently trust it because they're using it, you're much more likely, I think, to do that. And that that matters more so I think in this space than it has in other spaces previously. And so you asked like for specific use cases.

11:02 It feels like the thing and we can go in my my specific use cases in in a minute, but it feels like the things that you were seeing over and over again and the things that are clearly resonating with people is like look over either my expenses or look over my credit card bills or look over just a general bill of mine and find out a way to save me money, right? That's always going to be enticing to people, right?

11:23 like is there a way that is there something I'm not thinking about? like I have the XYZ credit cards and I'm paying XYZ bills like should I be using this credit card to pay this bill because they give you a discount or this one offers this service for free like and again there's been services that have done sorted these types of things before and and plugins and and browser plugins and whatnot but I feel like being able to speak in natural language and coming back to you in natural language not having to sign up for you know myriad new services to be able to do I think that was at least one of the early hooks that got people like, "Yeah, we're saving money.

12:01 This is great." >> Yeah. By the way, there's been some a lot of discussion including from some bankers about how like basically the banking industry is only solvent because people keep their money in like non-interest bearing accounts. And if these agents could efficiently move money from like cash accounts to safer investment accounts, you know, the the arbitrage that banks were doing for instance, you know, with cash sitting in accounts would cause a massive financial hit to the banking c to sorry to the banking industry. And not only that, you look at things like credit card points and sort of optimizing all these other purchases, loyalty cards. if you hand that over to these agents that upends so much.

12:46 >> Even worse than that, Alex is if just like and others have talked about this and I feel like this is this is quickly becoming our reality. Just imagine how many subscriptionbased businesses are based off of, you know, the fact that people just don't cancel, right? They they set it and forget it. And you know in the future if you have a an agent an AI agent who pings you every once in a while saying like hey I notice you haven't watched say HBO Max in a while or Netflix in a while like maybe you just want to like not not pay for it this month you know turn it off for this month and then if there's something great coming back on I can let you know and you can turn it back on. You know that's one very easy one. But even like any subscription service, right? It's like, yeah, you haven't read, you know, you subscribe to this this publication or this newslet >> it can it's a it's a problem for every for us, for all everyone else and YouTube and all of these things. It's this is a real real issue going forward and I don't know, you know, we're going to have to obviously have a whole new industry that sort of pops up around this and and figures out how to optimize these things and and keep people incentivized. Do we do more annual contracts or things and and do we have breakage fees and all that sort of stuff? But there's it's it's there's a lot coming with this.

14:04 >> You know, MG, I'm reading from your Yeah, there's there's still this sort of open question about whether these things are going to work or not. And I'm reading from the way that you're answering that you think that this is we've we've just seen the introduction of something that is going to be this concrete product that you know many people millions maybe billions of people are going to use. Is is your belief that this is not like flash in the pan type of product and something that has staying power? I do believe that at least one and probably a handful of these things will be used by most people going forward from here. I think there's a question of what the timetable is to get this to like mass consumer adoption and I do think that there is a lot of questions to be answered around some of the stuff we we're hitting on with like churn and whatnot because if if the service doesn't do something you know that that you're sort of looking for and and or you just more more honestly you just don't know what you should be asking of it because again a lot of people sign up for these things just to try it out. What's what's the downside of that? I'll just see what it can do.

15:07 But then after they do like the obvious stuff or the stuff that their friend told them, they're like, "Well, what else should I be doing with this?" And they just have no idea. And so, we can get to it in a second. But I think a lot of this will be prompting and and from what I've seen though in these early products. I think that they fully understand that. A lot of these companies fully understand that and they're doing some very smart things to sort of keep users engaged, keep people coming back to try it, to try new things, and to show real value.

15:35 including with myself. So so yes to answer your question directly I do think that these things are here to stay this time. I think that I don't know the exact time horizon for mass adoption like when our parents will be using them but I do think that they will be increasingly baked into sort of a lot of the AI services that we're already using. >> Okay. And you know my instinct initially no pun intended was to talk a little bit about each one one by one. but I'm going to shift gears a little bit and just ask you to break down the competitive balance and who has the best shot to win. So, just to talk through the players, we obviously have Instinct, we have Metamuse, we have OpenAI's new Dots product which they just released, we have Grockbot, we have a new product from Microsoft. let's not go too deep into Google Spark into Google Spark because we're going to talk about why Google missed this moment seemingly after the break. but of the players today, and I I don't know if I'm missing any, but those seem like the main ones. just break down like which personal assistant is going for what, and who do you think comes out on top at the end of this?

16:41 >> yeah. So, right now, I think, >> yeah, Instinct is was interesting as as sort of, you know, again, one of the early movers, first players here. and because they're a startup. Now, we talked about, you know, VCs putting putting out their their their tweets you know, to perhaps showcase how how much interest they have in the product. And, you know, by all accounts, that seems to have worked as it seems like by the reporting they've raised something like three rounds in, you know, a few months period. And they're maybe already valued at $10 billion.

17:13 >> Crazy. And their CEO is kind of >> the CEO is kind of like a relatively unknown person. young guy just kind of >> young guy came out of Sierra which you know full disclosure my family is an investor in and my wife is an adviser to as another AI startup but yeah he came out of that and we have no financial stake in in instinct but yeah so basically he obviously saw this world and and decided he needed to jump over and jump into this head first and so he did that and then you know, I think he also probably must have known because, you know, the rumors have been out there for a while that a lot of people, all the big players, which again we'll hit on in a second, we're all dancing around this space, and and potentially coming soon, with their own products. And so, Instinct is interesting still to me mainly in the the case that they have a shot at doing something interesting as like more of the Switzerland play here, right? because they're not backed by you know any of the major big tech companies right now. They're not a part of those. And so I feel like and this is sort of a different topic and and one that's already playing out in real time, but I feel like we're going to have these silos, these information silos that come into play more so than they have even with AI overall to date where certain players do not want their products, you know, being used by certain agents of other companies that they're rivals with, right? And so again, Instinct doesn't have that dynamic right now. Now they have the whole world of other issues like that are up against them, right? They don't even though they've raised a lot of money, they don't have billions and billions and billions of dollars to spend. They're still a startup. They're going to be beholden to other other models. They're going to be beholden to other clouds and all of and everything else. And then they don't have all the built-in mechanisms to to do distribution and all the other inherent advantages that the big players have.

19:13 So, you know, obviously >> Yeah. Sorry to pause you here and we're going to come back to it at the end of the break, but you talked about some players don't want their services plugging into others. We obviously saw that Amazon blocked Muse, so Metamuse can't shop on Amazon for you, but Shopify came in and said, "We'll work with you." But the one that's really interesting is, and I think you've written about this, and obviously we're going to get to it at the second, the second half, but it almost feels so important to insert now, is none of these things really work without Google.

19:42 like most pe I think most people well let's not I don't even want to say most or or less but we have billion billion plus maybe billions of people on Gmail Google calendar is obviously the biggest calendar when you I connect those two to these things they work when you don't connect those for most people they're not going to work and so it is interesting the dynamic here and I wonder if some of them are on borrowed time aka maybe Muse where Google which is typically pretty liberal in terms of who it allows to plug in with its services might just say, "Yeah, we're not doing that."

20:16 >> Yep. Maybe the biggest question sort of going forward and in the immediate future because you know basically Muse is obviously Meta's product right now. Dots is OpenAI's product. all of them are frenemies to some extent, right? They they either you know have compute deals or in in some cases they have other commerce related deals. even you know you mentioned Amazon blocking Muse you know it's interesting because Meta has other deals with Amazon in place you know for things for shopping elements and things like that and advertising for sure and yet they still like felt the need to do this to make that you know to to block that. And why is that? I mean obviously they have their own competitive products on Amazon. They have Alexa Plus now working. You know, they had Rufus, it was called, right, as their as their shopping assistant for a while, but I think they're transitioning over to to it being Alexa plus driven. And so, they want people to use their own thing. I think the bigger story >> have the advertising business around advertising. Exactly what I was >> Yeah, exactly. Okay.

21:22 >> Because they don't want other agents, you know, agents don't >> are not impacted by advertising that they see in their in their feeds based off of what they're searching for. And so, that, you know, potentially destroys what has been a huge growth driver, as you well know, for Amazon, you know, becoming one of the major players in advertising based off of that. And so, so they have all sorts of reasons for not not wanting to play ball here. But just to back up to go to go back to, you know, you asked like which one of the of the current sort of major players right now that have pushed their their agents and assistants out there. I feel like I was surprised honestly of of myself that I really like Muse Meta's product. I'm not a you know I am not a big Meta person. I've written a lot about sort of negative in recent years about them for obvious reasons but also like I I you know I've covered the company for 20ome years and I've gone through es and flows of where I thought that they were doing the right thing being from a product perspective let alone the the underlying fundamentals of the business and whatnot but I feel like you know I I do not like the state that say Instagram is in. I do not like the state that Facebook itself is in. They're all just, it just feels like all they're doing now is milking their billion plus user products to basically become, yeah, these ad vessels. and we're all just, you know, these these mindless meat meat people just sitting there looking at ads. And so I was surprised by how well and how well done Muse was from the get-go. And obviously, we've talked about as well the whole meta situation with AI. They had to do the big reboot.

22:59 I was pretty skeptical that they could sort of just buy their way back into the race and they obviously spent billions to do it from everything from doing the scale deal to building out more of their own infrastructure, doing a bunch of these other you know acquisitions and and and basically just getting up to speed very quickly over a year span and now to their credit not only getting models out the door getting a new product. This is not something that they've baked into Facebook or baked into Instagram. Now they probably will at some at some point but this is a standalone service. It has its own interesting security layer going on.

23:36 they built their own virtual you know private machines for you to use because I think they had to right like they had they knew that people are not going to trust this inherently something from meta especially something from meta and so I think that they did a lot of the leg work to get it done and their credit to their credit they got it out the door before even openi did and this is open's business AI you know they have no other business that they that they really need to focus on now they have other facets of the business obviously that they have been focused on but meta you know, the Meta AI team, the super intelligence lab, to their credit, they really, I feel like, knocked it out of the park with this early version, at least of what they launched with Muse.

24:15 >> So, you think that when we think about who's going to win that Meta has the best chance? Because I guess like there is this this this growing and you wrote about this a little bit about DOTS also, which is OpenAI's personal assistant. There is this growing divergence here between who which interests you serve, right? So Muse is clearly a consumer application. So I imagine a lot of people are going to be planning like vacations on it and maybe do using it to like look at their like their service subscriptions and figure out which one to cut. that seems like you know the use cases it will go for if people by the way this is I think a big caveat if people are willing to download the desktop app because there's the mobile app and the desktop app and from what I've exper experienced so far I don't think these things work without being on your desktop so I guess Muse will be the consumer application dots and you wrote about this last week despite the fact that Dots has this is openai's personal assistant has this like kind of like Sesame Street type feel with its mascots. It's an enterprise thing.

25:21 and so they're going after enterprise folks. I'm curious if you think are there going to be two winners here? One in enterprise, one in consumer. Does it all converge? what does what yeah I guess what does winning look like and who is the the ultimate winner? >> So and just to go back to one point you made about residing on the desktop. Totally agree. Though I also think that that Muse has an interesting advantage there in that from day one they they launched with WhatsApp integration, right? Something unique that Meta can offer because they own a WhatsApp. And so whereas we talked about Instinct sort of smartly leveraging chat, you know, iMessage and and other chat services. I think that Meta was very smart and launched day one with WhatsApp. So if you don't want to use you know a new service a new application, you don't need to. I mean I think that it's much more full fully functional and just has a pretty good UI to to sort of help you with what you can do if you do use the app. You can run it fully through WhatsApp and it totally works. Now, dots, open AI, >> is that 100% true? Because like you can run it through WhatsApp, but one of the nice things about dots is that it takes control of your whole control of your entire desktop. It takes control of your browser. You put a browser extension on, so it can go to websites and it can do things for you. And with I mean, I guess Muse has the virtual machine, right?

26:44 that it can use. But I think being in your context helps a lot if you just give it basically I don't I don't know if it's called root control or whatever it is and I know Apple is starting to have some reservations about this but just being in WhatsApp doesn't give you that full control and that I think is what you need to be make this entirely useful although it is scary to give that type of control to one of these things. So there's a couple things in that, right?

27:10 So you hit on Yes. So they have Muse launched again not talked about this a second ago, but Muse launched with basically a virtual computer in the cloud that is yours alone. and you know for each individual user and it's really a cool implementation down to the point where if you click on your muse and you can see basically you know when it's browsing a website and you can obviously take it over now other services have done that open had a version of that that you could do before but it really shows you that the full desktop is there and there was a there was a bunch of headlines around like people downloading their computer the contents of their computer in the cloud and people being a little freaked out by that and I thought the Meta's response was interesting to that which is like yeah that's the point that's it's your computer you can you can do what you want with it like it's yours you can download the contents you can do anything that's great and they've actually made it more explicitly like that now and I think OpenAI did as well they're even more leaning into that paradigm where you can click in and you can see like you can launch Chromium you know on their sort of virtual machine so there's that element of it which again I feel like is a smart thing to do It removes a lot of the worries about security and and some of this the concerns there. There's other levels of concern obviously that are baked into that still. But I did I think it was a smart thing to go with what you're talking about computer use.

28:37 Obviously a lot of these services have have had this for a while and it's it's to your point. It's why they needed to have not just web apps. It's why they needed, you know, native applications to be able to to port into Apple and and Microsoft different APIs on the operating system level to be able to control your computer. I would imagine that a lot at least Meta is thinking right now that they think that they can get done a lot of what people want to get done from an agentic perspective with this cloud service versus doing everything on your desktop. But you're right, like if you have an Excel sheet, you know, that that's not necessarily in the cloud or that just resides or that your your natural working, you know, relationship with it is on your desktop. Like there's obviously reasons why Anthropic has done super well with that. and so I think that there's anyway just laying out there's there's two different ideas there and how much that matters will ultimately determine like yeah which of these services ends up being sort of a winner if if any of them. Now you bring into that obviously Microsoft which is more recently come out with the the newly unified super app version of co-pilot and their play is much more of course down the consumer sorry down the enterprise front which is obviously their their bread and butter historically. They'll say they want consumer too, but you know, I think that they'll be happy as long if they can, you know, stay in the in the enterprise lane and sort of snuff out some of the wouldbe competition from OpenAI in particular because as you note, OpenAI's play here is much more sort of more business focused than what Meta is going after. Now, I do think again they all converge over time. They all want to own it all. They want to be the one agent. there is this open question and it sort of dates back to something I wrote several months ago ahead of all this agentic stuff which is the the notion of bring your own AI right like is is the workplace environment going to be such that you bring your own consumer AI service into the work or vice versa that you bring home your the the AI that you're already using like do people want two different AIs one for work one for home that may play out that way or it may play out that there's you know the one AI to rule them all of your life. And that's going to be a big question mark for this space in particular because again as you're noting like while Meta is going more straightforward after consumer and Microsoft goes more straightforward after enterprise, OpenAI sort of has a hybrid thing again with this with these cutesy figures which is just this weird dichotomy as I note in the piece between what that looks like and what it seems like they're trying to get you to do which is much more work oriented. I just felt like that that's like indicative of the whole company right now because obviously they've built this this consumerfacing behemoth in chat GBT itself. But then they of course had to sort of halt the brakes and and figure out like oh my god Anthropic has passed us in in terms of the business side and then in terms of valuation and like why is that? Maybe we should focus on what they're doing. And so to me that's sort of indicative of the whiplash the entire company has been under the past several months. And so I feel like DOTS is is in some ways playing that out in real time.

31:51 >> Yeah, OpenAI does feel like a little bit stuck between the two different use case. And it does feel like, you know, I guess it it I'd still say it's the maybe the best at product when it comes to AI. but it still feels like as opposed to setting the vision, it has been following to a degree. and the way that you describe these, I I do wonder like do they all blend in together? It's funny like when I was talking about computer use, you know, hearing your answer about Microsoft reminded me that Microsoft years ago had this product that like watches your desktop and gets stuff done for you and there was a backlash there and they ended up pausing that, right?

32:27 >> >> so just thinking through, we're going to have all these different players involved. so what is what would you say the differentiator is in terms of who's going to win? Is it going to be the one that like sort of nudges you the best? Is it going to be the one that's more effective in getting things done versus not? where do you think the line of of competition is drawn here and and who crosses the finish line? I do feel like there will be a huge first mover advantage to to a lot of these whoever can get sort of the most people signed up and actually using these products and obviously Meta has a huge advantage there because they have the multi-billion you know plus user base of several products and OpenAI has chat GBT which of course they baked this in to the Chat GBT super app from the get-go and so it's not like they they have nothing they have a billion Sam Alman said during the the dots unveiling, they're now at 1.2 billion weekly active users. So, it's not nothing to sneeze at, but it's not quite meta at the 3 billion plus, you know, user scale that they can get this in front of so many different services.

33:38 >> And OpenAI will will only let you use it if you're part of the pro subscription tier. They only use Dots. >> Good point. >> So, $100 a month. And I have to say, I just upgraded from 20 a month to 100 a month to be able to use Dots. I really like dots. I I mean, we spoke about this a little bit on the Friday show. I think it's great and you know, I know you got this feedback when you wrote about how much you like Muse. I just don't have any interest in letting Meta have access to Gmail and Calendar where for whatever reason. I wouldn't say I feel comfortable giving it to, you know, OpenAI, but I feel I guess a little bit less of a gag reflex when I connect it to Chat.

34:16 >> And I can't tell you how many messages I got along those lines. Now, I want to be mindful that, you know, we are we do what we do for a living. We're a bit in a tech, you know, bubble, you know, regardless of of where we live and whatnot. obviously, you know, the audience is is going to skew more in that direction than say the general public likely. but it is a real thing, right? Like there is this real concern that Meta is is just not to be trusted. And I think you know we're about to see it play out in Hollywood in real time with all these these movies coming out.

34:49 >> maybe this is maybe this is who determines who wins. We have an open AAI movie and a meta movie coming out and whichever one does a better job at portraying Sam or Mark Zuckerberg as I don't know a vile character. Well, they will face a harder time getting their personal assistant off the ground. I don't know. Is it as simple as that? And I believe and I think so this social reckoning which is the the meta the new meta movie sort of the unofficial sequel to the social network is coming out I think later this week and then I think artificial which is the open eye movie I think is having its premiere tonight.

35:25 so we're going to find out soon. And from but everything I've heard I I have some you know Hollywood folks I know and they're both portrayed insanely negative both Zuckerberg and Altman. So I don't know how this ends up. And by the way, there's also an Elon Musk documentary which is like four hours long coming out imminently and so he has his own AI service. >> When's the Dario movie coming out? Inquiring minds want to know.

35:49 >> We need we need the anthropic movie. >> That will happen by the way. I think you know the SNL thing you know that will happen. >> There's going to be the the effective altruism you know the movie and stuffed animals and things. but >> the theater production. >> Yeah. But yeah, that will be an interesting the reaction to those movies will be fascinating in the face of the trying to get these these adoption of these agents for the trust reasons that you're talking about. I mean, we joke about it, but that will be that will have some impact on like how these things are perceived in in public and you know, who wants to sign up for the meta version of a of a agent that controls your life versus who wants to sign up for the OpenAI version versus should anyone trust the anthropic version with all the stuff that they're saying about the end of the world.

36:34 Should anyone dare trust the Elon Musk version of this? Do we default back to Microsoft be because we should trust them or do we go with our old friend Google which I know you want to talk about in a second. And so yeah, this is this to go back to your initial question though of what ends up being the key things. So I do think that there's a first mover advantage here for sure which obviously hurts the the wouldbe startups. you know, there's Instinct, there's another one called Towns, there's a few other players that are I think there's a bunch more that are going to launch pretty soon. And so, they're unfortunately going to have a hard time sort of reaching scale as quickly as some of the other players that we're mentioning simply because of obviously the the baked in user install base. But there is an argument to be made for a pure play, you know, agentic service here. And again, the Switzerland argument, you stay away from all the infighting between these companies. Maybe there's an inroads there. But I do think it's probably going to be one of the big players. And I do think it's for ultimately also the reason that you hit on earlier, which is that the integrations are so key here. You need the integrations to mail, you need the integrations to calendar, you need the integrations to messaging. And which one of those players can nail all of that and keep them all intact importantly without being you know blocked and everything I think will be critical in terms of like how this ends up playing out >> on on the movies. I just want to say Sam Alman was on CNBC this past week and they asked him about it. Kate Rooney who did a great job with that interview asked him about it and he goes well looks like a terrible movie.

38:12 It's just the funniest response. He is also on the cover of Vanity Fair with some huge profile today. which I do not think is an accident in in face of this movie that's coming out. >> Oh yeah, there might be some coordination PR effort on that front. He did say something interesting in that interview I think or this was maybe the political interview where he was like opening eye and anthropic have fundamentally different worldviews on AI regulation and some AI benefit that that the AI benefits justify accepting some risks.

38:46 >> then there was no followup. Yeah. >> Yeah. >> Yeah. What what risks are we talking about here? right. Yeah. Yeah. >> I think it's Yeah. It's important to but let's see if we could channel him a little bit. I mean, I think I think what he would probably say is that there's going to be these sort of, you know, breakthroughs that will happen in the sciences and maths. We we've already seen the math breakthrough while Navier Stokes from OpenAI and probably more are on the way and that will help us understand the universe a little bit better, live longer, be more productive, be less tied to work, which is a big question mark. And the risks are, I don't know, cyber security probably, and what else would it be? I don't know.

39:27 Does that do you think that's the cost benefit that he's thinking through in his head or is there something else? >> Yes. I mean I saw the same the same quote and had the sort of the same reaction. I feel like to give him some credit I don't think that he's say suggesting that the end of the world is the downside you know scenario >> doesn't seem like it right. I think that he is saying that like he's trying to be realistic and you could argue whether or not that's smart, you know, in this in that in that type of response. But I do feel feel like he's trying to be realistic in suggesting that like yes, there are going to be downsides. He's talked about it before, right, a lot that obviously job displacements, you know, on down and and and sort of school displacement, all different sorts of, parts of life are going to be impacted by this in ways that are not universally positive. And so to suggest otherwise is obviously going to be disingenuous, but when you leave it that vague, it's, you know, it's going to again just lead to the continued messaging problem around everything that's been going on with AI. All right, MG, I I got to ask you on this one.

40:33 where is Anthropic here? I mean, they were effectively the they led this moment. >> It's a good question. Yeah, >> they had cloud code and co-work initial let me give control to an AI and see what happens. and >> people took to it and they became the world's hottest AI company >> because they built those tools and now >> maybe they've decided that this is not a place they want to play in or they've decided, you know, you know, sort of in disagreement with us on this show that this is not going to be a popular product category. What do you think their their perspective is and why they're absent here?

41:12 >> yeah, it's a it's a good question for all the reasons that you noted. It's something I've been thinking about in the back of my head as well. and I did I haven't explicitly written about it other than to mention them that they don't have, you know, a straightforward play here. They might it might be as simple as they view that that basically, you know, Claude Code and more specifically Claude Co-work, I guess, are will evolve into this, you know, as they add more features over time. but you're right, they were the first mover with with, you know, sort of the coding movement. Now, you could say you know that that that there were others who who sort of actually showcased it a little bit to them. but they obviously came in and that's like what's what we were talking about has risen the company you know to the to the levels that they're at right now and then co-work you know opening eye everyone else felt like they had to have an answer for that and so where are they here? To me, it goes back to what I thought about before when they were still very much the underdog in the space, which is that they've always sort of despite hiring people like Mike Kger. they've always sort of shied away from being more consumer oriented and and sort of always wanted to be more about, yeah, work and and and sort of productivity and whatnot. And so I feel like they their answer to that would probably be like, look, we want to do whatever yeah, helps our our user base be more productive, but we're not necessarily looking to you know, to do all of the consumer use cases. Remember, they still don't even have an image generator product. they have like sort of minimal voice capabilities. I do think that they'll have to ramp that up more, I think, because voice will become a more key integral point of how you interact with these services over time.

42:55 But again, they haven't done a an image product. They haven't done a video product. And you know, in some ways, looking back, you could say that was smart, right? They weren't distracted by those things that perhaps OpenAI was. But I also think like an image product is probably table stakes at some point too, right? Because you're making a presentation. are you going to use another services sort of you know image generating you know model to to go along with the the presentation that it's making like there's stuff that they will have to do and I do feel like that eventually they will be probably dragged into this more even if they view it more consumerf facing personal AI assistant mode. I would imagine though they'll say like look we're going to start with with a professional AI assistance and we already have a lot of that with with co-work thanks to all the work that we've done with cloud code and and so we'll you know we'll get to that when you know down the road and and we'll sort of morph into that.

43:51 They did do an interesting thing which you probably saw I think it was just last week which is that they merged the claude and co-work textbox. So they're no longer the two, you know, the two different tabs that you click between, which even OpenAI has not done yet. And so I feel like they're moving towards the the holy grail of that space, which is the unified single chat box for anything. And I do feel like DOTs in some ways might be pointing at this as well. Like is DOTS actually going to be the eventual total front end of chat GBT? and it it uses its stuff behind the scenes to basically determine what you're asking for, whether it's a personal question or a a work oriented question and roots it behind the scenes for you.

44:39 >> Yeah, I could see it. I mean, one of the nice things about dots is you can like sort of continue to throw stuff at it and then it will basically prioritize and work behind the scenes and then execute for you. So, all right, we've kind of danced around it through the show. I will give you my perspective on who I think will win this battle and my dark horse and I'd be curious if you have an answer on your end. I think OpenAI is going to win. I just think they have shown that they have the product chops. They have 1.2 billion users. So, it's not like they're in Instinct where they're like trying to fend off Meta's incredibly aggressive distribution, attempt, which is not only pushing it, Meta's not only pushing this in its own products like WhatsApp and Instagram, but it's also it's got the that's got ads during NFL football. It's everywhere in the sub. I'm in the subway. All I see is Muse. so I I think Open AI does have that 1.2 billion users, people who are there to use AI, people who are more comfortable to use its products. And I think DOTs is going to be the one that comes out on top. And my dark horse here, of course, wouldn't be a show without between us without bringing up Apple. I think Apple, which will have device access, from the get-go and has shown an interest in being more aggressive on AI products, is the dark horse to pull this off, you know, even if it might be last. What do you think?

45:57 >> I I feel like I'm going to sound like that that one scene in Jurassic Park where I'm like the and and I'm the one defending the blood sucking lawyer. you know here where it's like I'm the one defending Meta. Can't believe I'm the one defending Meta. but I I feel like one other element that we haven't really even hit on, which is interesting, is that Meta is doing all of this without what is, you know, what they even consider to be a frontier model behind the scenes powering this, right? Like obviously Muse Spark, I think it's 1.3 now. It might be 1.4. I think it's 1.3 is good. and it is powering, you know, this right now, powering Muse right now. But they say any day now that quote unquote watermelon, their their actual big time, big frontier version of of their Spark models or sorry, their Muse models, Spark is is one variant of it are is coming soon.

46:55 And so when that happens, what does that change? Does that change anything? I mean like presumably it will be much better much more functional at doing certain tasks but maybe they built Muse in such a way that it is fine just running on you know not really good but not quite frontier and maybe you don't need it for for all of these agentic tasks and maybe that maybe we we're starting to break into a world where the absolute frontier matters for science and for you know all of these really hard discovery and and big data stuff. but sort of the day-to-day operations can sort of fall back to these models which of course every on top of everyone's mind is is that they can run cheaper, they can be offered cheaper. and and so that's that's the world in which we're playing. But again, I just want to put that out there that Meta has been able to do this without having the absolute frontier model whereas obviously OpenAI does right now and Anthropic as we talked about is not playing in the space right now but has that model. And so, I'm I'm curious to see what, if anything, changes when they launch watermelon. Do they launch with new functionality for Muse that just blows people's minds? Maybe not.

48:05 Maybe it's not. >> That's Meta's big upcoming model that we're we're waiting for. >> Yeah. >> Yeah. >> and then you bring up Apple. so again, I wrote a post right after WWDC, after they announc, you know, launched the new Siri AI officially, at least unveiled it and now it's out there in the world. I still feel like that that post holds up where it's like Apple could win and probably will win AI by default just because again it's baked into billions of devices automatically. I do think that there's an interesting element which I think I hit upon a little bit back in that piece and I know that others other you know key Apple watchers like John Gruber and others have been talking about more recently which is that the biggest problem with Siri AI and and Jason Snell I know has talked about this a bit too is that you just forget that it's actually good now and so you forget to use it right even though it's baked into your phone it's like it's right there everyone has it you don't need to download a new app but you're like still opening up chatbt and opening up claude queries that you could easily do on the in this bakedin version of Siri and you just forget that you can trust it now. And so in some ways it has the opposite trust of what meta what we were talking about with Meta, right? It's just you don't trust that it's going to with Apple. You don't trust that Siri is going to work because it hasn't for over a decade. And now the fact that it does is good obviously, but it's still a problem now because they have that trust deficit in another weird way. And so I do think like time will alleviate that and especially as they release new devices. you know the rumors are that they're going to have their first sort of like you know home device that could be largely predicated around Siri AI. It sounds like they're working on getting new HomePods out there imminently. And so I do feel like that will help a lot of these problems. But you're right the key thing that Apple has which no one else has is again they're on these devices. They have inherently baked in support for mail, messaging, your ca your calendar, your pictures, >> see everything on your screen.

50:03 >> They they can theyact they can say you can say, "Hey Siri, what am I looking at right now?" Like you can't do that with the other services. So, unless you, yeah, move heaven and earth to to basically agree to all of these prompts, which as I think you alluded to earlier, Apple is moving to potentially lock that down more, at least on the Mac, in the future. They haven't exactly explicitly stated what they're going to do, but they're clearly going to put up more warnings that even the Mac, which historically has been open to whatever, you know, you as a user want to do to it, you know, will potentially now be getting these these prompts like, "Hey, maybe you don't want to give Meta access to all of the information on your machine."

50:44 >> I wonder if it' be more effective if they would have said, "Maybe you don't want to give Meta all this access or Mark Zuckerberg." I feel like maybe again after the movie, we'll see. Did you see the movie? You know what this person wants. >> Now free on Apple TV Plus. >> Exact. >> It's just a giant ad for streaming. >> got to ask, did you did you do you have the Fold? Did you buy the Fold the Duo?

51:10 >> I have not. It's up for pre-order on Friday. >> Oh, okay. >> Have it. Yes. Yeah. >> Too early. >> I will pre-order it. I will yeah, get it then. I think it's the week later. so I have just I I broke down and bought an 18 Pro just to have something, you know, while I wait. >> All right, we'll get your review in in our November show. So, let's take a break and come back and talk about the thing that I've really this this whole conversation I've been looking forward to, but I've kind of been chopping at the bit and barely holding back from asking you why has Google not gotten into this battle. So, let's do that when we come back right after this.

51:47 And we're back here on Big Technology Podcast with MG Seagler of Spyglass. You can find Spyglass at spyglass.org. Highly recommend you sign up for the newsletter. One of my favorites. all right, MG. So, let's talk a little bit about Google, shall we? >> let's do it. >> Google always seems to be a little bit a little bit far behind on these on these product releases and that is also the case here. they have this AI assistant called Spark but certainly not in the conversation with the others.

52:18 And like we mentioned in the first half or in the beginning because we only have a couple more minutes left. they have Gmail, they have calendar, they have talent, they don't have a muse, they don't have a dots, they don't have an instinct. What is going on there? >> And now I'll do another movie call back of Swingers. If you remember the scene where it's like you're the bear, you're the bear and you got these claws and you got everything. You you just got to you got to use these claws, man. that is Google.

52:46 >> that's the bear. No claws, though. What's going on? >> They're they're Yeah, they're defanged. the problem is, you know, and again, I should disclose like I normally do. I work there for over a decade, but I haven't been there in a couple years at this point. And nor did I work on AI specifically for the company. So, I don't have any real insight insider insight into this beyond you know, everything that I that I read and think about. And that is you know what we've talked about over the past couple of years. They were behind. They they moved heaven and earth it felt like to sort of catch back up. and I think that they did that for a little bit and then lo and behold they it seemingly fell behind again. And then you had the new shakeup of the company on the AI side of the deep mind side. And is this all related? I mean to some extent for sure like there's elements of it that are that are definitely related. In my head and I wrote a piece about it. It just goes to the idea that clearly IO this past IO la a few months ago they go out on stage and they just don't have anything really ready to go. They have some little products here and there, and they're willing to talk about like these things that are coming in the future, but they sort of screwed themselves by, you know, like pre-announcing these things, including famously 3.5 Pro, their forthcoming flagship AI model, which never came and now seems like it's never going to come.

54:08 I think they've actually acknowledged that now because they have their first version of Gemini 4 out in the wild for security testing and and whatnot. And so I think that part of it was probably they also at IO talked about a product which is also confusingly called Spark but not to be confused with Meta Spark but it's another AI Spark Gemini Spark which is their version of you know Agentic and Assistant use cases and they had it in beta testing but it was very limited. It was only I think at the ultra tier at one point. I think it's come down. I've tried it. I've had access to it for a little while. It's fine, but it's not, you know, it's not blow away blow your socks off AI.

54:51 It does certain things well because as you note, they have the integrations to Gmail, to Google Calendar, to everything else to Google search. Like, they should be the ones, they have the clause, they should be the ones to do this. And yet, they cannot seemingly time it right. They can't get out of their own way. I mean, I've always had problems with Google from a product perspective, the way that they implement a lot of things. Obviously, you know, you it's hard to argue with with multiple products that have over a billion users, but I would still argue with a lot of them. I think that they could be a lot better in terms of functionality and in terms of like user user interface intuitiveness.

55:30 And I feel like that's it's pretty key in these early days for a product like this in the AI personal assistant space. And and again, Gemini Spark, it's not doing much for me. I think it's not doing much for a lot of people, especially when you compare it to to Muse and now Dots and Instinct and everything else that's out there. >> Yeah, you're right. Muse is full of fun and whimsy from the Avatar on down.

55:54 Gemini Spark lacks that. What's the word I'm looking for here? >> Yes, it's a little on the nose, but it's it's right there. It's like that's Google, though. They just again, I'll keep reiterating it. They have the claws. They have everything they need. They have the talents for sure. And they had I mean the Deep Mind Purchase will be one of the greatest acquisitions of all times. And they have everything that they needed to be to be in the right place, right time. And even when they fell behind, it felt like they again they did everything right to come back and catch up and be there. And it felt like we remember we were having the discussion is it over? Is like Google going to run away with the game? And then all of a sudden now we're having the opposite discussion again. And it's >> it's like they've done nothing since that conversation.

56:37 >> It's like Groundhog Day and and you know they have they have reasons for all of it and they'll they'll say some of it you know externally even now because they sort of have to of like you know I think Sundar you know said like yeah we sort of I don't know if he said drop the ball on coding but he said that they weren't focused on it as they should have been appropriately and then so that was obviously a problem for a while and then again it feels like there was a similar idea. They just have way too much going on. This is what it always comes back to for me. They have too many cooks in the kitchen. They're they have way too many people that work there that all want to work on the same thing, but all for different reasons don't want to work on the same team. So, they have like all these conflicting cooks, all these conflicting you know, motives and and it's not all nefarious or or you know, mean-spirited. I just feel like there's different teams. It's a massive company. you know, the the Google Cloud team has different incentives, right, than than the the Android team and the Google Search team has a different incentive than all of them. And so, it's very hard to get all those people to work together to get on the same page.

57:39 And I say all that and I but I thought we were over that when we had the unification of Google Brain and Deep Mind and that that was going to be the time and it felt like it was working. And then all of a sudden though these these products creep up and you know I talked about in that piece there's a product called CC which by all accounts sounds interesting right it's like meant for families but when you have that product and you're trying to do to be a family based assistant when you have that product and then you have spark >> you have spark and you also point out search with AI mode >> right all the all the AI agentic stuff built into it like if you have everything people are going to use nothing because like they don't know which one to go to like for this and so you can make an individual case that all these are the right calls or they're nice products or you know they do this and they do that but they need to get on the same page and again they thought it felt like they were doing that you hear all the reports Sergey Brin involved like that they're they're really honing in and they just can't get out of their own way for whatever reason.

58:41 >> Yeah. And this is sort of the point where we have these conversations and I'm like, okay, just like, you know, remember, I'm telling myself, remember be that, you know, it'll probably just be another month and they'll be on top again the way that this stuff has been. >> That's probably true, right? They'll they'll have Gemini 4 out there for consumers eventually and then they'll they'll presumably probably try to tie that to a hopefully they'll rebrand it, but the Gemini Spark, you know, full-on launch and they'll go for it and say like, "Hey, you saw we you saw all that cool stuff you could do with Muse and Dots will look at what you can do with all the Google services that you already use like and you know, basically combat it that way and it will probably be compelling for a little bit. But then now you have another we go back to the trust notion. you have the the trust notion of like should I put my entire agentic life and the personal assistant in Google's hands when they're just going to fall behind again in six months.

59:34 >> I mean they really and I know this is a sort of a you know beaten down trope that's been going on for a while. They need to just like acquire a creative agency and then get get this together in terms of like their messaging and product names because it's I don't understand how you have this problem persist for so long. It's it's mindboggling. and speaking of acquisitions, maybe to tie it all back into the beginning point is you know, I would expect that someone will acquire Instinct, someone will acquire maybe the town's product, someone will acquire all these other things basically because you know they'll feel like look we are we're seeing what what Meta has done. We're seeing what OpenAI has done. we need to be in this fast. Like Microsoft, we'll see how that goes. They obviously launched theirs. Where's Amazon? Right.

60:21 Amazon has had Alexa plus out there for a while now. why where are they in this space? There's been rumors that they have you know work behind the scenes on a more agentic workflows for this kind of use cases. And so where are they in this? what does SpaceX SpaceX SI now it's called? I guess >> Oh my goodness. Don't get me. We have another hour. >> Sexy. yeah. and so where where do they take >> Oh my god.

60:47 >> Where did they take Grockbot? because Grockbots's gotten some some love out there, right? It's >> right. We didn't even mention Grockbot. I mean, >> I know it's I think they're doing some smart things with Teams and and some of the others are having to combat that. and so they've done good work, too. so yeah, there's >> it's it's yet another battle. >> The website spyglass.org. MG Seagler joins us on the first Monday of every month. MG, always great to speak with you. Thanks so much for coming on.

61:14 >> Thanks, Alex. Talk soon. >> All right, everybody. Thanks so much for listening and watching and we'll see you next time here on Big Technology podcast.