[00:00:00] You’re listening to a new episode of The Brave Technologist. This one features Ahad Mato, who is the founder of Intelligent Internet, as is dedicated to advancing, distributed and accessible artificial intelligence systems. As a founder of Stability ai, he influenced the widespread availability of generative AI tools for image, audio, and video reaching over 300 million downloads.

In this episode, we discussed their vision for an open and intelligent internet and how it will impact the everyday user every day. What problem agents are likely to resolve today, and his expected timelines for future unlocks. Specific examples of how they imagine agents safely and intentionally removing cognitive friction for users, the incoming low cost of intellectual work and impact on the everyday user and advice for parents raising a next generation of users.

And now for this week’s episode of the Brave Technologist.

[00:01:00] Amad, welcome to the Brave Technologist. How are you doing today? I’m doing very well. Thanks for having me on. Thanks for coming on. I’ve been looking forward to this one. you’ve kind of described intelligent internet as a mission to bring the best open intelligence to everyone.

What does open intelligence mean to you in practice and, and what does a user first version of that future look like from your point of view? So I was thinking a lot in my previous company, stability, ai, we’re building all the image, video, audio kind of technologies about. What does the AI of the future look like?

And it was like, it’ll probably look like Jarvis for Ironman, right? You come in and you talk to your AI and it also talks to your kids and it runs the governments and things like that. And I was like, that first AI that’s next to you should probably not be run by someone else and working for someone else, you know?

Mm-hmm. Like it needs to be open. The AI that, you know, teaches your kids, manages your health guides, your government should be open source fully, and it should be aligned in a way that’s transparent. So its only objective should be your flourishing, [00:02:00] your community’s flourishing, your government’s flourishing.

So I think in practice, what it looks like is you’ve got agents that then become actual buddies. But their guts and the way that they’re trained is fully transparent. It’s fully open and it’s given as a human right. And if you can either use decentralized version or you can run it yourself if you want.

And again, that’s at the individual level, up to the state level. Awesome. That’s great. What kind of gaps or mistakes from the early generative AI wave are you specifically trying to correct with the intelligent internet? So I think that early wave we were trying to build a GI and these very generalist systems, right?

And so we were training on massive amounts of data, but there was so much junk in there. you had some weird things like, you know, the whole trademark dispute issue. There was just, uh. victory for stability AI at the high court in the uk. ‘cause it was like, oh, it generates watermarks sometimes. Like we didn’t want it to generate watermarks and like that’s bad data.

Just like I don’t want my medical AI to have any [00:03:00] Reddit knowledge in there. You know, that’s gonna be bad data. I think that the focus on billing generalized polymath agents got away from like, what do you actually want the AI that teaches your kids to know? What do you want it to the doctor to know? And the way that you build these types of AI is almost like these.

Civic ais, these civil servant ais, is very different to building the generalized polymathic ais. And right now you’re trying to take the polymaths and turn them into doctors or teachers or things like that. Whereas I think you can build much more resilient, robust versions of that using the right types of data in the right way, which also requires an order of magnitude less compute, or actually a couple of order of magnitudes, less compute, both to create and run.

Yeah, I, I think you know, it’s uh, an interesting perspective kind of approach. And I love the kind of the openness around it. And, and so you all are releasing kind of everything from like a medical reasoning assistant to like a collaborative agent hub. How do you decide what problems agents are actually ready to solve responsibly today?

So I think that [00:04:00] you’re moving from this era of prompt based stuff, which is like a really smart guy. You tap on the shoulder and you say, blah, and he says blah back, and gives you a poem or an image or something like that to these agents that can work on very long periods of time. That are more economic in the way that they work in that individual tasks don’t have a big economic impact, but being able to make whole movies or hold reports and then iterate back and forth and be proactive, that’s definitely economically valuable.

So our thing was, let’s build the best generalist agent and then specialize it. So we just released i I agent a few days ago. It’s number one on terminal bench two, it outperforms cognition and others and it’s fully open source and you can use your clawed code subscription or whatever for it. But now we’re gonna specialize that into.

Basically. Doctor agents, nurse agents, teacher agents, PA agents, policy agents. We just announced something with Saudi Arabia and FII, which is a policy agent for the world so that you can dynamically create national policies, compare them against constitutions and pledges and other things. And [00:05:00] see what the latest impacts of blockchain to quantum, to AI are on the way that you should model your nation.

So these real specialists, ai, social scientists, as it were, as opposed to kind of, AI scientists or mathematicians, et cetera. And again, the way that we use those is just very, very personal ai. And that’s gonna be a release in Q1 that you can just talk to and it’s fully open source that you can run on your own hardware.

You can run with any model or you can use the hosted version that’s basically delivered as a base level free of charge. So we’re working on how to do that right now. That’s awesome because I feel like, you know, just looking at kind of one, there’s just this, all this hype around genic and when you see a lot of it out there kind of in practice, a lot of it seems kind of, you know, par or tricky or like, oh, you’re just showing me how you can do something that I can do faster than that.

but it seems like you guys are kind of looking into these these very specific kind of, you know, specialized areas that, that. Could be meaningful and, and useful, which is awesome to [00:06:00] see. you just mentioned too, you touched on blockchain a minute ago. You said AI agents in crypto can kind of belong together for the average user, what becomes possible only when agents are paired with like decentralized identity or payments or compute.

So I think it’s the value exchange on your behalf, right? So like for example, you know with the Brave Browser you had the B token and it was like there’s an advertising component and then a match component, and you’ve been thinking about the way that looks. That looks very different as brave becomes more and more agent and can make micro transactions legally as well, which is also important.

Other things, tell me about it on your behalf using X 4 0 2 with the tax code done properly, et cetera. Like it just unlocks this whole world because typically we have intents and we have actions. Mm-hmm. But there’s a lot of friction between them, right? Mm-hmm. Because a lot of our intents are quite fuzzy.

Mm-hmm. And the logic flows weren’t enough to unlock that. Like you can have a system that [00:07:00] learns the type of thing you’d like to look at. And then adapts to that to show you the content that you want and then pay people properly for that content. Similarly, we’re gonna outsource more and more of our mental minds and our shopping, and more to these AI agents that operate on our behalf and interact with other AI agents.

And so the rails for that need to be interoperable. They need to be dynamic low cost, and crypto’s fantastic for that. Mm-hmm. The other aspect of crypto is it’s verifiability. And this coordination layer. And so for what we’re doing, the Civic ai, we created a version of Bitcoin that we call Foundation Coin, whereby it’s just on high performance Blackwell, GPUs and equivalent.

And we stack those, we sell coins, and then it goes a hundred percent into a healthcare supercomputer for cancer and you know where the coins are going, or it goes into a computer to do the culture of Illinois. And free AI for the people of Illinois. Mm-hmm. So it’s like Bitcoin, 21 million supply, [00:08:00] except for every primary coin sale, helps people building trust and network effects.

And we thought that was the most elegant way to do this. Yeah. ‘cause the amount of compute and the use of this AI is actually inevitable. So when you’re trying to build corporate AI that you’re selling to people, you don’t know if it’s gonna be used or not. Right. Mm-hmm. But I know that in 10 years, every single child in America or the UK will be taught by an ai and that AI should be self-sovereign.

I know every government will be guided by an ai, and every health system will be run by an ai. What if it was run and securing a Bitcoin type currency that will be more distributed than Bitcoin. Every coin sale goes to good, and we stack the GPUs for good. So innovative mechanisms like this are also unlocked by this technology, which I think will be super cool.

Yeah, I think so too. And I think it too, another point is it is just the ability to do things, you know, like cross border with crypto is just like really awesome. And you know, removes the friction part too. I mean, like, and, and even [00:09:00] getting back to, to kind of, you know, our model and brave, I mean, the way AI is kind of disrupting things, it’s really interesting to see because even the way that people think about advertising, like.

Your brand will always be, need to be known, but a lot of the kind of hook in and distract to get you into this thing will kind of go away when you have an agent that’s you know, learning off of what you’re doing. So I totally agree. I think it’s really interesting way of looking at, like you said, it’s the right thing at the right time, and it’s about these agents remove cognitive friction, right?

the danger here is again, like is it operating on your behalf or someone else’s behalf? And so I think the agent that’s closest to you, like I don’t want my kid being taught by chat GPT, but I’m fine with the agent next to her using chat GPT, which is like, let’s build this protocol that gives everyone universal AI and build civic ai and then use that to secure compute that secures this currency.

That we can then sell to get more compute for this. That’s the feedback loop that we had. Mm-hmm. For intelligent internet. And then if we release it fully open source, anyone can [00:10:00] use it if they don’t want to be a part of that network. ‘cause permissionless innovation, like when we release stable diffusion is such a key core part of this as well.

Yeah. No, that’s awesome. I think openness is key. And and, and speaking on this, you know, kind of this area what, what’s the biggest misconception from your point of view that, that people have about a genetic AI and crypto that kind of prevent them from seeing the natural connection? So, I think, you know, you, you haven’t really seen the big use cases ‘cause crypto’s been legal pretty much, and it’s about to turn legal because like you needed to build.

You needed to build the AI first, and then you build the protocols and the crypto. So like things like X 4 0 2 for stablecoin payments, microtransactions have just come around, so we’re setting the building blocks there, but agents are literally only a few months old. I would say like the task that an AI could do six months ago was a minute at most, now it’s like seven hours, right?

Right. You’ve got memory, you have [00:11:00] ontech sharing, you have the ability to do automatic verification. Proactive work that’s all coming together right now. So a lot of the Crypto XAI protocols kind of jumped the gun a bit and they had bad token economics or they didn’t round this ‘cause they went crypto first versus AI first.

But now you’re moving to a point where the amount of agents is gonna go from thousands to millions to billions to trillions, and you need a coordination layer for that. You need standardized protocols for that. You need interoperability with MCPA two A, et cetera, and that’s all coming together as of next year.

So I think that No, that’s awesome. It was just a bit early, the initial ramp. Now it’s calmed down and, but next year’s gonna be massive across the board with all of this and, and with these things, I mean, this is something we kind of, are doing. I mean, as we, we look at adding a genetic. You know, to the browser aside from the security concerns, right?

And kind of like doing it in an isolating, you know, or doing profile based things and making sure that, you know, [00:12:00] things are safe. it is interesting, like, and I’d love to get your take on this, like just trying to kind of figure out. What age genic use cases people are gonna find interesting at the current stage.

Right? Because, you know, people will say, okay, shopping or I’ll have it write my emails or tweet from me. But like, it seems like too, part of the ego wants to kind of do that manually too. It’s like, oh, what, what, where’d you, what’s your kind of take on, on the current state of like useful genic.

Capabilities for people you know, in the, in the current, in this current moment, moment where, where do you think there’s gonna be the most usefulness? as people kind of start to adopt this stuff? So you think about the browser. The browser and information augmentation and coordination mechanism, right?

But it’s very forgetful and lossy. Like you go back and look at your history every so often. You do clippings, et cetera. The nice thing about having an in place ai. And again, this is why they’re also trying to dominate with Atlas and comment, and everyone’s trying to own that user interaction layer is that it could act proactively to surface stuff that you might like.

It can track your projects, it can [00:13:00] do these analysis, and then it can say, for example, I could do another level of analysis, but it will cost y amount of compute, you know? Mm-hmm. Okay. Or it’ll cost Z amount of research access. We’re a microtransaction to Springer or this or that. I think that’s what we’ll see because again, people do use browsers for shopping, but they use it for information and knowledge and research primarily, as you said, like you might still wanna buy your own thing.

Does it really save that much to go and click the buy button? On a discretionary purchase. Not really. And like I said, that’s fine, but do I really want to do my groceries? Not really. You know, like Right, exactly. Exactly. The the day-to-day stuff I want my assistant to do and the browser ai. Is pretty much an assistant that’s the age agentic nature, and the best assistant are proactive.

So I think that what we’ll see is the browsers kind of evolve to more project and data structure based things whereby something like a deep research, for example, you know, you go [00:14:00] on to chat, GPT, use deep research, you get a thing and it’s like 70 pages long and you read it and you’re like, okay, that seems interesting.

But then you don’t really go back and edit it. It, right? Like it should be a project in your browser that gets dynamically updated regularly, and then it can say, I can look at these different things and here are different budgets of my time to be able to do that. And you’re like, yeah, this is really great.

I need to super polish it. And then it calls a Figma agent to make the design look even nicer, right? Mm-hmm. Or I want to have some in-depth market analysis. It calls a Bloomberg agent. To get all the latest data and make it up to date and stuff like that. So in that way it becomes more this control plane layer of information value transfer, and he said then it can do the day-to-day tasks of calendars, et cetera.

But then the discretionary stuff is more like knowledge share understanding. And they said humans like to click the button on those. Yeah, exactly. Or, or see the difference in the sparkly laces versus the none. [00:15:00] Right. Like, or, or whatever in, in this types of thing. But, but if you’re looking at drops of new sneakers, you want your browser attached.

You say, Hey, look, you spent like this much time looking at these sneakers. Look what’s coming. Right. That that’s there. I want to be notified of that type of stuff. Well, here’s a new game that’s like’s. Yeah, but it’s very different if it’s like meta recommending that to you versus it is verifiably acting on your behalf.

Like the advertiser can still play for that. See for that click through, right? Mm-hmm. But what you want to know is that the AI is credibly neutral or aligned with you. Yeah. Versus working. No, it’s a huge difference and I think this, again is the really exciting thing about crypto, about the advances we have in ZK now and other things like that.

The work that Egen layer has been doing with Egen Cloud, et cetera, or near with intense, like you can say like This browser is acting on my behalf. [00:16:00] As it gets more and more intelligence around it, it’s surfacing stuff that, yeah, we can have transparency over the payments. They are being connected to me, but the value is then mutual as opposed to I feel like a beer.

Yeah. It’s gonna be a Bud Light always, because you know, they’re maxing that out. Exactly. No, that’s, that’s great. if you had to kind of explain the intelligent internet vision to someone who’d never touched an LLM or wallet, like how would you paint the picture of an Asian enabled world?

Yeah, so I think, again, the way that I put it is. AI is gonna teach your kids, manage your health, guide your government, who’s building that and who owns it, you know? Yeah. And so the way we’re doing that is we’re making sure that AI, as it gets smart and it can talk to you and engage with you, and grow with you, is gonna be owned by you and working on your behalf.

And it’s gonna be made available to everyone. ‘cause it’s the biggest uplift we can possibly give humanity. And I, know you mentioned policy earlier, how has it been, have you been working with like, you know, governments on these issues and, and how has it, how has the response been [00:17:00] generally, you know, working with people in, governments or on the policy side?

So we, we launched, like I said, the SAGE initiative with the Kingdom of Saudi Arabia, their big conference, FII for this AI that can take the latest knowledge on quantum blockchain, everything, and automatically create dynamic policy, allow you to compare blockchain standards in your country across others.

Eventually allow people to give their input on policies and co-create policies, and then everything can be done transparently. Check if things are constitutional governments love it. And so like, again, they’re all scared about what’s coming because so many technologies from quantum to ai, zk are all maturing at the same time.

Robots are coming and they’re only gonna cost a dollar an hour, you know, like. Big upheavals are coming, and this is where I wrote a book about this, the Last Economy. We need something to help guide through this that is credibly neutral and that’s the idea of sage top down. The other part of it we’re doing is universal ai.

Bottom up again, how do you give everyone in Pakistan, for example, it’s just on their government or [00:18:00] Rwanda, et cetera, or even Illinois. Or Utah and AI that’s aligned with them as this control plane. You know, they can still access other ais, but you know, is on their behalf and we figured out mechanisms for that.

And again, governments are hyper engaged with that. So we’re talking to lots of them and we’ll start rolling it out in Q1 probably of next year. Oh, cool. Very cool. Yeah. ‘cause it seems like, you know, for, for the lay person kind of looking into this, you typically just see all the noise about jobs or environmental stuff and things like that.

But I, I feel like people don’t really aren’t really as aware of how engaged a lot of the policy side is with the people building and stuff. So it is really cool to hear that perspective on it. Well, they don’t have any alternatives as the other thing that we found, like, you know, you look at the Trump tariff announcements that was clearly calculated by chat, GPT.

Someone figured it out from the country codes and you’re like, I remember seeing some stuff about that. Like GPT-4 Oh. Did our national tariff policy. No, that’s not good. Right.[00:19:00]

I think you know, and you touched on this a little bit earlier too, around, around costs getting cheaper, you know, and, and, and if we get into this era where kind of intellectual work becomes abundant and, and nearly free what new measures of human value do you think we will rally around? So I wrote this book, the Last Economy, where I discussed that.

And I said that the way that mathematically, actually, so we have a whole math paper that’s coming out about this right now. We look at material, so you know, that’s GDP, but the AI will outperform us on GDP ai. First companies will be human companies, robot first companies will beat human companies, you know.

We need to look at the other things. And we found the other key measures you need to look at is more applicative are intelligence. So that’s how knowledgeable your society, your community, your group, your yourself, is the network effects. So there’s a huge amount of brand value for something like Taylor Swift, it isn’t quality, it’s brand value, right?

And then diversity for resilience. And there’s a whole bunch of metrics you can do to measure that. Because if you’ve got a really [00:20:00] great community and a really great family, you are less likely to be impacted by disruption than if you don’t, if you’re all alone. Right. Mm-hmm. Mm-hmm. If you have lot adversity and you have lots of different jobs from agrarian to white collar, again, you’ll be less disrupted if you have intelligence capability, which in this case is actually how to use the ais and being like a first user of them.

So you’re using rep plate or P suno or I a agent daily, you’ll do far better in what’s coming from a capability than anyone else. Mm-hmm. So we suggest a whole bunch of different measures for these, and ultimately, I think. Long term, just a couple of decades from now. ‘cause most people, even the bears are like a GI in 10 years.

We’re, we’re, we’re like we have to look and articulate a Star Trek type future ‘cause there’s more than enough food in the world to feed everyone. But we lack coordination. There’s more than enough to give everyone housing. You can give everyone a young Lady’s illustrated primer and intelligence just for them.

You know, you can do proactive healthcare [00:21:00] for everyone for pennies a year. These are fantastic things. Mm-hmm. But the transition period is gonna be crazy and a lot of jobs are gonna go and they’re not gonna come back because the AI can do it literally for pennies. Like, actually, I’ll give you an example.

Grok 4.1 Fast came out yesterday, right? Yeah. Yeah. And they had, it scores like 95% on taube. So above even Gemini two point. Gemini three Pro Taube is basically customer service, telecom agent benchmark. It’s wild. And so, a human is like 80 or 90%, but Grok 4.1 fast is 50 cents for a million words. Oh my gosh, it’s 20 times cheaper than sonnet.

And so you look at that and you’re like, okay, that’s like an entire call center replaced for like 10 bucks, you know? Yeah, yeah, yeah. And then you consider like how many tokens? It’s like not point, it’s like 1.3 tokens per [00:22:00] word are needed to do your ta, your crypto taxes. Right, right. It’ll probably be like a buck.

To do that in a few years. Right, right. Versus right now it’s almost impossible to do crypto taxes. You look at how many tokens it is to build an app. That’s a good one. The average app is a hundred thousand to 200,000 words of code that’s about a buck in a year or two. Mm-hmm. To build that end to end, and you’re like.

This seems weird. It’s like, yeah. Yeah. it brings me to kind of another thought too. I mean, and, and I’m seeing this, even like my, physical therapist came back, he’s like, yeah, I was to a conference. And they have this AI that’ll basically do front desk for me now and it’s specialized in, trained on physical therapy stuff like databases and, and all these types of things.

Like, I’m really curious to get your take, you know, when you step back and you see how. Your neighbors or your friends, or your even non-technical people are adopting these tools now. Like is there any surprises from your point of view? Are you, you expect it to be kind of like how you’re seeing it [00:23:00] or were there any things where you’re like, oh, I didn’t necessarily anticipate that, or, you know, any, any stuff like that?

Just curious. I mean, I was always bullish on this, you know, and I thought that again, this would be used everywhere. And I left stability my previous company about a year and a half ago. ‘cause I was like. This is the takeoff point now where you’re gonna go from prompting to every single one of these technologies is gonna satisfy.

So it’ll be good enough, fast enough, and cheap enough. Mm-hmm. My shocks were that people were still using GPT-4 Oh. And they didn’t know about thinking and things like that. Right. But then I was, yeah. Yeah. Well that’s only, it’s only be a couple of months. People are adopting this even faster than I thought they would.

Mm-hmm. And again, I think the next stage is when we get proper agents in the first half of next year. Mm-hmm. And you’ll be able to talk to them just like. A human, you’ll have a zoom call with them or you’ll give them a call or whatever, and then they’ll just go and do work and there’ll be swarms of them.

They don’t have the capacity, they just spin up more. And so I think, you know, like the current one is as I expected, I thought there’d be maybe a bit more drop loss. Now, we haven’t really seen it in the numbers apart from in the early [00:24:00] stage, but I’m even more confident now that we’re gonna see a big disruption coming next year, the year after, and definitely three years from now.

But the technology’s progressing as I thought it would, but it’s, I mean, like Nano Banana three came out yesterday, right? I, yeah, yeah. I knew it was coming, but the fact that it could take this entire podcast and one shot an infographic of it. It is wild. It’s wild. It’s, it is so wild. I, I just am, am blown away.

Like in the past year even, you know, just looking back at like how clunky stuff was with the image side of this and, and how smooth it is now. It’s just incredible. And two years ago this week we released stable video, which was state of the art open video model.

It was like, it would take you going from like static like this. To gentle movements like that, you know? Right, right, right. And now you’ve got so too, you’ve got Cling, you’ve got all these other things. You got your Hollywood level, and next year it will be real time. [00:25:00] Next year is real time, 4K video generation.

You can feel it getting closer and closer. I mean, I feel like it’s just yeah, it’s incredible. I mean, and, so just kind of like, zooming out a little bit more too, like if I was a parent of a very young child, you know, what should I encourage them to learn to help them with their future career?

They should be using notebook halem, and repli and suno and all these things every single day for an hour. Ideally with other people and just learning to create. Mm-hmm. Like again, the capability shifts are just crazy. Like, again, let, let’s look at Gemini three Pro. Mm-hmm. AI generated websites looked purple and weird.

Until Gemini three Pro came out and now you see all these people doing like particle effects and all sorts of crazy cool stuff, like if you use it daily, then you know as soon as this stuff comes, but like you should be encouraging them to build. Like right now, if you start on video, I guarantee that if you have a team of young people coming together, you [00:26:00] could build a feature length movie over the next year.

That is as good as anything you see on tv. Mm-hmm. Using different in the right way. Yeah. But that experience of doing that will be huge. Like why don’t you just do it as a family, you know? Right. That’s what I encourage people to do. Like spend an hour a week with your family building stuff. It’s the difference between everyone sits down for dinner to everyone makes dinner together.

Right. Like it, it is so funny that you mentioned that too. ‘cause I’ve been doing this with my daughter where I’ll draw a sketch, right? and I’ve got these like pasta markers, which are these fantastic. If you’ve never tried ’em, try ’em out. They’re, they’re wonderful. but you kind of make the sketch and then Yeah, I was just throwing it and imagine, right?

and then I take it to my daughter and be like, how do we wanna storyboard this out? And she would tell me the next steps, the next steps, the next steps. And then all of a sudden we’re like, at the point where we’ve got enough of these clips to stitch ’em together into like a. A 32nd or a minute movie, right?

it’s super fun, you know, like, or bedtime stories or things like that, you know? Yeah. That’s one of the cool things that you can do is, so if you’ve got an image that you do, [00:27:00] you tell Nano Banana to annotate the movements on it, and then you put that into video or whatever, so it draws the lion actually on there for what’s the most natural movement.

Oh, I’ll definitely try that out. That sounds really cool. I mean, this next question is kind of silly ‘cause I feel like I’ve heard like 10 bold predictions that actually sound pragmatic and real. But, but you know, if there’s one bold prediction kind of about the future that you’d be willing to bet on today, what would be Oh, bold prediction on the future?

I, I think that in within two years, you can generate the entire Game of Thrones, season eight, but good. Within two years. Bold indeed. Bold indeed. Maybe they can finish the wins a winner while they’re at it. I mean, let’s not go too far. Let’s not go too far. Awesome. Awesome. Well, hey Matt, I mean, this has been really a fantastic discussion.

I really appreciate you making the time. If people want to follow your work where should they go? And go to ii inc. You know, please do download our book. It’s free [00:28:00] in the Last Economy, and leave some reviews, uhhuh and I’m at EMO stack on Twitter as well. Fantastic, Ahmad, I, I, again, thank you so much for, for making it out today.

Just a fantastic conversation and love to have you back sometime too to kind of, get an update and maybe the, maybe the podcast interview will, will interview itself like by then. Who knows? I’ll send my digital twin next time, unless I’ve already done that. Yeah, I’ll be talking to your agent, you know, right on a mud.

But thanks so much for dropping by and and, and have a great one. Yeah. Cheers. Thanks for listening to the Brave Technologist Podcast. To never miss an episode, make sure you hit follow in your podcast app. If you haven’t already made the switch to the Brave Browser, you can download it for free today@brave.com and start using Brave Search, which enables you to search the web privately.

Brave also shields you from the ads trackers and other creepy stuff following you across the web.