Vana Cofounder and CEO Anna Kazlauskas speaks at Imagination In Action’s ‘Forging the Future of Business with AI’ Summit about how Vana helps people collectivize their data.
Subscribe to FORBES: https://www.youtube.com/user/Forbes?sub_confirmation=1
Fuel your success with Forbes. Gain unlimited access to premium journalism, including breaking news, groundbreaking in-depth reported stories, daily digests and more. Plus, members get a front-row seat at members-only events with leading thinkers and doers, access to premium video that can help you get ahead, an ad-light experience, early access to select products including NFT drops and more:
https://account.forbes.com/membership/?utm_source=youtube&utm_medium=display&utm_campaign=growth_non-sub_paid_subscribe_ytdescript
Stay Connected
Forbes newsletters: https://newsletters.editorial.forbes.com
Forbes on Facebook: http://fb.com/forbes
Forbes Video on Twitter: http://www.twitter.com/forbes
Forbes Video on Instagram: http://instagram.com/forbes
More From Forbes: http://forbes.com
Forbes covers the intersection of entrepreneurship, wealth, technology, business and lifestyle with a focus on people and success.
Subscribe to FORBES: https://www.youtube.com/user/Forbes?sub_confirmation=1
Fuel your success with Forbes. Gain unlimited access to premium journalism, including breaking news, groundbreaking in-depth reported stories, daily digests and more. Plus, members get a front-row seat at members-only events with leading thinkers and doers, access to premium video that can help you get ahead, an ad-light experience, early access to select products including NFT drops and more:
https://account.forbes.com/membership/?utm_source=youtube&utm_medium=display&utm_campaign=growth_non-sub_paid_subscribe_ytdescript
Stay Connected
Forbes newsletters: https://newsletters.editorial.forbes.com
Forbes on Facebook: http://fb.com/forbes
Forbes Video on Twitter: http://www.twitter.com/forbes
Forbes Video on Instagram: http://instagram.com/forbes
More From Forbes: http://forbes.com
Forbes covers the intersection of entrepreneurship, wealth, technology, business and lifestyle with a focus on people and success.
Category
🤖
TechTranscript
00:00 I wanted to start with a throwback from my time at MIT.
00:06 So I came to MIT obsessed with central banks.
00:10 I had had a picture of Janet Yellen in my high school bedroom.
00:15 Worked at the Fed just like obsessed because I thought currency was the central thing.
00:20 And anyway, I came to MIT and found that you were actually able to run basically a decentralized
00:29 central bank from your dorm room.
00:31 So I would go to the loading docks, find old GPUs and computers that labs had gotten rid
00:37 of and set them up to mine Ethereum.
00:41 Because essentially what it was doing is like shifting -- using technology to shift power
00:47 to the people, right?
00:48 And so how do you take decentralization and allow it to create this distributed central
00:55 bank?
00:56 And in the case of Bitcoin and Ethereum, they've done that really well for finance and currency.
01:02 In the case of VANA, we're doing that for data and AI.
01:06 So how do you put the power of AI and data in the hands of the people?
01:12 With VANA, your data and your AI is fully yours.
01:16 So it's self-hosted, it's portable, and it's monetizable.
01:20 Just to get a sense of the room, how many of you have run a machine learning model from
01:24 your computer or other hardware you control?
01:27 Okay, great.
01:29 So today, usually when you use a service, you have to kind of put your data inside of
01:33 that service, right?
01:34 So there's this double-spend problem where any application you use, you have to trust
01:39 with your data.
01:41 What we've developed at VANA is essentially this personal server architecture where your
01:46 data and your models live in an environment that you can control.
01:51 So right now, I have my data hosted on my MacBook so you can have a variety of different
01:57 data sources.
01:58 You're able to export this from any platform because of data regulation.
02:03 And then you can run a local language model alongside it.
02:07 And so here I've got my personal server running from my MacBook.
02:10 We have around a million people who have their personal models stored in VANA.
02:14 Most people do not run it from their MacBook.
02:16 They work with a trusted provider such that it's a little bit easier.
02:20 But what I'm able to do then is choose where it lives and then make it portable, right?
02:25 Like there's not that much you can do with a model locally.
02:28 You want to use an application.
02:30 So here what I'll do is bring my model over to an application called uSlap, which will
02:37 give me daily motivation in the sound of my own voice based on whatever I want to work
02:43 on.
02:45 Cool.
02:47 So what's happening here is I'm granting access to this application to talk to my personal
02:52 server.
02:53 And now I'm going to tell it what I want to do.
02:55 So what am I currently working on?
02:57 I want to improve my daily meditation practice.
03:00 I've been traveling a lot and I've kind of fallen out of it.
03:03 Daily meditation practice.
03:06 And my AI model also has a lot of information on me.
03:11 So it's going to personalize this and be able to use so much context on me while still preserving
03:17 my privacy.
03:18 Okay.
03:19 There we go.
03:20 >> Embrace the tranquility within, even amidst the bustling digital world.
03:30 Consistent meditation practice fortifies your emotional equilibrium and nourishes your intellect
03:35 like a soothing balm on the soul.
03:40 Allow this quiet space to fuel your creativity and expand your consciousness.
03:45 >> So this particular application leaves you a voicemail every day in your voice kind of
03:52 hyping you up.
03:53 And it's a really early example of the sorts of applications that are possible.
03:57 One, it would be really scary if someone could take this clone of my voice and use it for
04:02 whatever they want.
04:03 And so that's why it's so important that you own your AI and your data that allows you
04:08 to create things like this.
04:10 So I was just showing you that portability aspect that VANA enables.
04:14 The third aspect that I wanted to talk about is actually monetization.
04:18 So today we're in this time where there's a huge economic shift and people are really
04:23 worried like, hey, is AI going to take my job?
04:27 And what's kind of happening is a lot of big technology companies are taking data from
04:33 all across the internet.
04:34 A lot of it is probably your data if you post on the public internet.
04:38 And then training an AI model that can do your job and replace you.
04:42 And it's a really bad deal for society.
04:46 Now there's sort of a race to get the private data that's not available publicly.
04:50 So you see things like Reddit is selling data for $60 million, et cetera.
04:57 And if I imagine a world where there are like ten different AI Annas that can be autonomously
05:03 doing work and earning money, it's very important that I'm the one who actually owns that and
05:08 who gets to decide, like, hey, what are the rules?
05:10 What can this do?
05:11 What can this not do?
05:13 And so with VANA we've let people actually collectivize around their Reddit data.
05:19 And we've created the first Reddit data collective, which allows you to export your data and then
05:25 put it into essentially a place where it's held in escrow with everybody else's data
05:30 and monetize it on your own.
05:34 And this sort of decentralized AI ecosystem enables you not just to own your own AI, but
05:40 also the AI that gets collectively created.
05:42 So you can imagine a user-owned foundation model created by 100 million people who all
05:48 contribute their data.
05:50 They can also train a small piece of the model locally and use some techniques like model
05:54 merging to then have a better model.
05:57 Today open source AI kind of eats the leftovers of big tech.
06:01 We get a model that's a few generations behind.
06:04 But really people can actually create an even better model if we're able to collectivize
06:09 in the right way.
06:10 Thank you.
06:11 [ Applause ]
06:11 >> All right.
06:12 So that's it.
06:13 I'm going to go ahead and close out the session.
06:14 I'm going to go ahead and close out the session.
06:15 I'm going to go ahead and close out the session.
06:16 I'm going to go ahead and close out the session.
06:17 I'm going to go ahead and close out the session.
06:18 I'm going to go ahead and close out the session.
06:19 I'm going to go ahead and close out the session.
06:20 I'm going to go ahead and close out the session.
06:21 I'm going to go ahead and close out the session.
06:22 I'm going to go ahead and close out the session.
06:23 I'm going to go ahead and close out the session.
06:24 I'm going to go ahead and close out the session.
06:25 I'm going to go ahead and close out the session.
06:26 I'm going to go ahead and close out the session.
06:27 I'm going to go ahead and close out the session.