Latest / Crypto Coin Show / AI data is centralized and expensive — Perceptron built the decentralized mesh that cuts costs by 92%. Peter Anthony, Co-Founder
Transcript
- 0:00Speaker 1: I'm Ashton Addison from the Crypto Coin Show and today
- 0:03Speaker 1: on Blockchain interviews with Peter Anthony, co founder of Perceptron Network.
- 0:07Speaker 1: A high quality AI training data which can get information
- 0:12Speaker 1: at a fraction of the cost of what centralized providers charge.
- 0:15Speaker 2: And also you're having.
- 0:17Speaker 1: That ability to earn from data that you're contributing and
- 0:20Speaker 1: running through a decentralized network also, which is super exciting.
- 0:24Speaker 1: Excited to dive into this, Peter. Data is so important
- 0:27Speaker 1: with the growth of AI. The biggest companies in the
- 0:30Speaker 1: world are paying lots of money to get high quality data,
- 0:34Speaker 1: and people that contribute that data should be able to
- 0:37Speaker 1: earn from it. So thank you so much for taking
- 0:40Speaker 1: the time.
- 0:40Speaker 2: To come on the show. Thank you for having me.
- 0:43Speaker 3: Looking forward to talking about data in the world of AI.
- 0:46Speaker 1: Definitely, it's always nice when you can earn, especially if
- 0:49Speaker 1: you're sitting on AI all day. It's back in before
- 0:54Speaker 1: Web three, people were getting sort of tired of Web
- 0:57Speaker 1: two and being like, well, you know, I sort of
- 0:58Speaker 1: feel like I am the product. I feel like that's
- 1:01Speaker 1: more so the case even with AI. On these centralized platforms,
- 1:06Speaker 1: everything you type and respond to is used to make
- 1:10Speaker 1: their platforms better, and you have to pay them for it,
- 1:13Speaker 1: not really earning anything.
- 1:16Speaker 3: Absolutely, I mean every single minute of every day that
- 1:19Speaker 3: you and I or everyone out there who's using a
- 1:22Speaker 3: mobile phone or laptop but contributing about one point three
- 1:25Speaker 3: megabytes per second of information. Now, a lot of the
- 1:30Speaker 3: data that's in existence right now has already been straped
- 1:34Speaker 3: by the existing AI models, so Open AI, Clawed, Anthropic, etc.
- 1:39Speaker 3: They've all been able to gather the existing data. That's
- 1:42Speaker 3: something like seven hundred and fifty trillion tokens of data.
- 1:46Speaker 3: Token is effectively three quarters of a word. Now, that's
- 1:49Speaker 3: going through all the forums, going through allus on X,
- 1:53Speaker 3: going through all the things on.
- 1:54Speaker 2: Quorra and Reddit whatever else.
- 1:57Speaker 3: And that gives you, obviously a portion of the Internet
- 2:00Speaker 3: portion of information, but that's only a fraction of really
- 2:03Speaker 3: the kind of information that's available. The majority of information
- 2:06Speaker 3: that is useful for training the future of AI is
- 2:09Speaker 3: what you and I are doing on a daily basis,
- 2:11Speaker 3: how we're interacting with claud, how we're interacting with Twitter,
- 2:16Speaker 3: whatever that might be. Now, yes we're contributing or yes,
- 2:19Speaker 3: we're producing data, a lot of that data is just
- 2:22Speaker 3: going to waste. What is missing is perhaps being able
- 2:26Speaker 3: to gather that data and make use of it and
- 2:29Speaker 3: reward the people who are contributing that data.
- 2:32Speaker 2: Definitely.
- 2:32Speaker 1: I saw that Google and OpenAI had paid multiple hundreds
- 2:37Speaker 1: of millions to Reddit and to get access to some
- 2:39Speaker 1: of these parts of the Internet where people are contributing
- 2:42Speaker 1: their own information. I'm guessing none of that actually went
- 2:45Speaker 1: to the users of Reddit. It'd be nice to have
- 2:48Speaker 1: like a little payout, but I feel like that's something
- 2:51Speaker 1: that needs to be fixed, and decentralized networks should have
- 2:55Speaker 1: that capability of contribute contribution and then earning from it.
- 3:00Speaker 1: So is that how perceptual network is laid out.
- 3:04Speaker 3: Yeah, So at the moment you and I, when we're
- 3:07Speaker 3: contributing data that's being used from Google or whoever else,
- 3:11Speaker 3: obviously they're packaging that data up and selling it on
- 3:13Speaker 3: for millions, if not hundreds of millions of dollars to
- 3:17Speaker 3: other companies. Now, what perceptual network allows you, as an
- 3:20Speaker 3: individual user to do is to be able to run
- 3:23Speaker 3: a node. Now that node can be run on a
- 3:25Speaker 3: computer or on a Android device, and this is effectively
- 3:29Speaker 3: you sharing your excess bandwidth. It's giving us a advantage
- 3:35Speaker 3: point of the Internet. It's not us gathering your data,
- 3:37Speaker 3: but it's us being able to look at your access
- 3:43Speaker 3: point of the Internet, your IP address, to be able
- 3:46Speaker 3: to gather a package of data. Now we are able
- 3:49Speaker 3: to do this across our network. We have around eight
- 3:51Speaker 3: hundred thousand downloaded nodes. Each and every one of those
- 3:56Speaker 3: nodes gives us advantage point where we can gather pieces
- 3:58Speaker 3: of packets of data and build them into meaningful data sets.
- 4:02Speaker 3: And by running a node, you will earn You'll be
- 4:04Speaker 3: able to earn points. Points in the future will translate
- 4:08Speaker 3: across to perceptual tokens. In addition to that, when you
- 4:13Speaker 3: gather data, you know, raw data as as it is,
- 4:17Speaker 3: isn't always as valuable as it can be. But you
- 4:20Speaker 3: can add value to data by building it into meaningful
- 4:23Speaker 3: data sets. You can annotate different data's pieces, et cetera,
- 4:27Speaker 3: which kind of upscales the value. Kind of like if
- 4:30Speaker 3: you think about milk. You know, milk's very cheap, but
- 4:33Speaker 3: if you go and buy cheese and ice cream or
- 4:34Speaker 3: whatever else, it's the same product, but it's.
- 4:36Speaker 2: Kind of made more expensive.
- 4:38Speaker 3: By kind of adding annotations and creating more value in
- 4:42Speaker 3: the data, you can create more value from the end
- 4:46Speaker 3: crime and what an individual user in our network can
- 4:49Speaker 3: do is they can contribute not only by running a
- 4:52Speaker 3: node to contribute to the bandwidth, but they were also
- 4:54Speaker 3: able to participate in things like annotation tasks as well
- 5:01Speaker 3: as contributing to data sets. So perhaps we need pictures
- 5:04Speaker 3: of certain types of people, perhaps we need to have
- 5:08Speaker 3: images of certain places in certain cities of the world,
- 5:12Speaker 3: whatever it might be. You can then go out in
- 5:14Speaker 3: your city, you can take photos, you can contribute to
- 5:16Speaker 3: data sets, and ultimately you can then earn rewards for
- 5:19Speaker 3: these different data tasks.
- 5:22Speaker 1: That's very cool. And so from what I understand, there's
- 5:25Speaker 1: sort of a passive part and an active part. You know,
- 5:28Speaker 1: if you are a busy person. I think everybody has
- 5:31Speaker 1: excess bandwidth on their phone or the computer they're not using,
- 5:36Speaker 1: especially at nighttime, unless you're running AI agents twenty four
- 5:39Speaker 1: to seven at maximum capacity maybe the point one percent.
- 5:43Speaker 1: But you can run a node and not have to
- 5:46Speaker 1: actually be actively contributing the valuable part of data. It
- 5:50Speaker 1: still provides information with earning potential. But then if you
- 5:55Speaker 1: go full active mode and you actually contribute the exact
- 5:59Speaker 1: type of data that these companies or perceptional network is
- 6:02Speaker 1: looking for then you boost your earning potential. Is that
- 6:05Speaker 1: is that right? Yeah?
- 6:07Speaker 3: Effectively, effectively that it's you know, like like anything in
- 6:10Speaker 3: the world, the more difficult it is to deliver a
- 6:13Speaker 3: particular thing, obviously, you know the value of that becomes more,
- 6:17Speaker 3: becomes higher. But there is obviously the passive kind of
- 6:20Speaker 3: contribution that people can do by running by running nodes
- 6:23Speaker 3: and effectively, what we do is we take data. We
- 6:27Speaker 3: then sell it on to clients. Obviously clients who are
- 6:30Speaker 3: building things like AI models. We also gather gather kind
- 6:35Speaker 3: of like live data sources, sentiment analysis, particularly in crypto.
- 6:39Speaker 2: It's a very useful thing.
- 6:40Speaker 3: Can then be sold onto exchanges, quant desks, et cetera.
- 6:44Speaker 2: And this then you know, creates revenue.
- 6:47Speaker 3: Revenue then goes back into gets distributed amongst the contributors,
- 6:53Speaker 3: but also kind of goes to developing future products and
- 6:56Speaker 3: building up the building up the business as a whole
- 6:59Speaker 3: to you know, on board more clients, you know, get
- 7:02Speaker 3: the whole ball rolling.
- 7:03Speaker 1: Definitely, well, it sounds enticing, especially with this job market.
- 7:07Speaker 1: People are always looking for side hustles or you know,
- 7:11Speaker 1: extra income. But of course there's probably a lot of
- 7:14Speaker 1: variables in how much you can actually earn based on
- 7:18Speaker 1: the passive methods, and then actually how what kind of
- 7:21Speaker 1: data you're contributing that's needed. How does that mechanism work
- 7:26Speaker 1: in determining how much potentially you can earn from contributing
- 7:31Speaker 1: the data?
- 7:33Speaker 2: Good question.
- 7:34Speaker 3: At the moment, as we move towards completion of our
- 7:37Speaker 3: data questing app, there's going to be certain data sets
- 7:42Speaker 3: that are obviously much easier to contribute to because you know,
- 7:46Speaker 3: if we are English speaking, for example, we need something
- 7:49Speaker 3: written or sorry, read out in the English, it's very
- 7:52Speaker 3: easy tasked. Lots of people are gonna be able to
- 7:53Speaker 3: do it. It's a lower value task. Let's say, however,
- 7:56Speaker 3: that we need a rare language from somewhere sub Saharan Africa,
- 8:01Speaker 3: and maybe we've only got five or ten users on
- 8:03Speaker 3: our network who are able to actually read out that
- 8:07Speaker 3: particular passage rare example, but that would obviously command perhaps
- 8:14Speaker 3: a slightly higher reward for doing such a process. The
- 8:20Speaker 3: mechanism via which points to things that gathered is obviously
- 8:26Speaker 3: a complex one, and obviously it's quite heavily regulated by AI.
- 8:30Speaker 2: But we've just acquired.
- 8:34Speaker 3: A new company, oh sorry, well, an older company in
- 8:37Speaker 3: crypto called router, and this company has effectively built a
- 8:41Speaker 3: framework that allows us to identify each and every user
- 8:45Speaker 3: and how they've been contributing to the network.
- 8:47Speaker 2: So it's effectively done automatically. No one's really sort of
- 8:49Speaker 2: sitting down and working out individual contributions.
- 8:53Speaker 1: Wow, that's really cool, and I'm sure there'd be some
- 8:58Speaker 1: those few sub Saharan Africans would be more than willing
- 9:02Speaker 1: to contribute some of that rare information.
- 9:06Speaker 2: I would love.
- 9:07Speaker 1: To dive into that app and see what are the
- 9:09Speaker 1: things right now that my expertise can can fill.
- 9:12Speaker 2: In those gaps.
- 9:13Speaker 1: So is that with the earning potential? You said right
- 9:17Speaker 1: now you can earn points and then later that moves
- 9:20Speaker 1: into tokens. So can people go to the website to
- 9:24Speaker 1: start or is it Android and iPhone applications?
- 9:29Speaker 3: So the best thing to do right now is if
- 9:30Speaker 3: you go across to Twitter. In fact, actually if you
- 9:34Speaker 3: go to the website, which is cetrons with an S
- 9:38Speaker 3: dot xyz over there, you're going to be able to
- 9:40Speaker 3: download the application, either for a Chrome browser like I'm using,
- 9:45Speaker 3: or for Android device, iOS device, etc. So once you've
- 9:50Speaker 3: kind of downloaded it eventually right now, because we are
- 9:53Speaker 3: pre tge, you're going to be earning points in the
- 9:56Speaker 3: future once the token has become live. Those points can
- 10:00Speaker 3: can translate to being worth token to And obviously as
- 10:05Speaker 3: the network grows and as the network brings on more
- 10:08Speaker 3: paying clients, that gets distributed throughout the network.
- 10:12Speaker 1: Yeah, it always pays to be early on, you know,
- 10:14Speaker 1: the next big thing. So I can leave those links
- 10:17Speaker 1: in the show notes below some of the points that
- 10:22Speaker 1: I was looking through on perceptions. It was saying that,
- 10:24Speaker 1: you know, the cost is much lower to acquire this information,
- 10:28Speaker 1: and I'm curious, how does that economics work versus centralized
- 10:35Speaker 1: AI collectors. You know, they paid two hundred million for
- 10:38Speaker 1: Reddit information, would they pay an equal amount for Perceptron data?
- 10:46Speaker 3: But right now, like effectively, one of the major problems
- 10:51Speaker 3: in AI is the monopolization of access to data, compute power,
- 10:57Speaker 3: et cetera.
- 10:58Speaker 2: And this is predominantly due to cost.
- 11:01Speaker 3: Now, one of the major highlights, or one of the
- 11:04Speaker 3: major things that's highlighted the actual.
- 11:05Speaker 2: Problems of monernopilization happened the other day.
- 11:10Speaker 3: US based Anthropic, which is where you run claud etc.
- 11:15Speaker 3: Was allowing people to use this new model called fable
- 11:18Speaker 3: five and then all of a sudden, it shut down
- 11:21Speaker 3: access to fable five for everyone outside of the United States.
- 11:25Speaker 3: Now that's great if you're in America and the problem okay,
- 11:29Speaker 3: But if you're outside of the United States, it highlights
- 11:31Speaker 3: that there's a company out there that's heavily relied upon
- 11:34Speaker 3: by a lot of different people that suddenly stopped giving.
- 11:38Speaker 2: Access to people who might need it.
- 11:40Speaker 3: That can have a really big impact in the future
- 11:42Speaker 3: because it dictates these people are going to be allowed
- 11:46Speaker 3: to do building and whatever else, but these people are
- 11:48Speaker 3: not going to be able to That's a huge problem
- 11:51Speaker 3: with monopolization. Another problem with minipolization when it comes to
- 11:55Speaker 3: data is the access to data and how you do it. Now,
- 12:00Speaker 3: traditional methods to being able to gather data has come
- 12:04Speaker 3: through access to APIs for Twitter, for Reddit, etc. Now,
- 12:08Speaker 3: obviously the data is extremely valuable, and because it's extremely available,
- 12:13Speaker 3: if you own those companies, you can charge huge amounts
- 12:15Speaker 3: of money for it.
- 12:16Speaker 2: Now, the problem with doing that is if you charge huge.
- 12:18Speaker 3: Amounts of money, only open AI and Anthropic and a
- 12:22Speaker 3: handful of others are going to be able to afford
- 12:23Speaker 3: to pay those data costs. If you are a new
- 12:26Speaker 3: upcoming AI project, you're never going to be able to
- 12:29Speaker 3: access that quality of data.
- 12:31Speaker 2: Therefore, your products are not going to be very good.
- 12:33Speaker 3: You could build the best products in the world, but
- 12:35Speaker 3: it doesn't have access to the data.
- 12:37Speaker 2: It's not really very useful.
- 12:38Speaker 3: Okay, so it's like you could be the smartest person ever,
- 12:41Speaker 3: but if you've never read a book, you're not going
- 12:42Speaker 3: to know much.
- 12:44Speaker 2: Right. So what we do that's different is that.
- 12:48Speaker 3: We look through the eyes of an individual user and
- 12:53Speaker 3: we do that hundreds of thousands of times, which enables
- 12:56Speaker 3: us to gather the data and information that would be
- 12:59Speaker 3: show through APIs.
- 13:01Speaker 2: Now.
- 13:02Speaker 3: This is something that recently went through court and has
- 13:06Speaker 3: been identified as legal in that it is id access
- 13:11Speaker 3: in the Internet without any.
- 13:14Speaker 2: Having to go through pay walls, et cetera.
- 13:16Speaker 3: So if right now you're able to access Twitter or
- 13:19Speaker 3: Reddit from where you are in the world, then I
- 13:22Speaker 3: can also access Twitter or Reddit from where you are
- 13:24Speaker 3: in the world, and I can do that hundreds of
- 13:26Speaker 3: thousands of times over to be able to gather the
- 13:28Speaker 3: same meaningful data sets that any of these other companies
- 13:31Speaker 3: can do, and I can do it for a fraction
- 13:32Speaker 3: of the cost. Because I can do it, I can
- 13:36Speaker 3: gather the data for effectively free and then obviously, you know,
- 13:40Speaker 3: we make we need money to build network, to run
- 13:43Speaker 3: the network.
- 13:44Speaker 2: To distribute to users, et cetera.
- 13:47Speaker 3: So we're breaking down the monopoly effectively by being decentralized.
- 13:51Speaker 2: Yeah.
- 13:52Speaker 1: No, that's one of the most important factors of decentralization
- 13:56Speaker 1: is global access and equality for everyone in the world.
- 14:02Speaker 1: You know, one thing I'm curious about with regards to
- 14:06Speaker 1: the data quality, you know, quality of data is super important,
- 14:10Speaker 1: probably much more so than quantity. If there's all of
- 14:14Speaker 1: these data questing tasks in Perceptron and I go in
- 14:18Speaker 1: there and I'm thinking in my head, I'm just going
- 14:21Speaker 1: to do all of this as fast as I can.
- 14:22Speaker 1: I don't really care about the quality myself, as long
- 14:25Speaker 1: as I put stuff in there, get paid, and get out.
- 14:29Speaker 1: How does the network, you know, and the AI mechanisms
- 14:33Speaker 1: behind that determine the quality of data and ensuring that
- 14:37Speaker 1: you're actually getting paid for putting in quality information and
- 14:40Speaker 1: not just junk.
- 14:42Speaker 2: Good questions.
- 14:43Speaker 3: So, I actually one that I was kind of talking
- 14:46Speaker 3: to someone about today. First of all, we've been looking
- 14:50Speaker 3: at this with regards to the data questing platform and
- 14:53Speaker 3: how to effectively identify you know, serious contributor versus a
- 14:58Speaker 3: non serious contributor. Non serious contributors who perhaps are not
- 15:03Speaker 3: you know, we're trying to kind of do every different
- 15:06Speaker 3: tasks that are not necessarily doing it properly, and they're
- 15:09Speaker 3: quite clearly not able to do the task they want.
- 15:11Speaker 3: We'll obviously have some sort of a penalty for not
- 15:16Speaker 3: contributing data effectively, but also we have right.
- 15:21Speaker 2: Now and what we use.
- 15:23Speaker 3: So for example, when we scrape, let's say someone wants
- 15:25Speaker 3: to go and get us.
- 15:27Speaker 2: Someone's requested that we want to.
- 15:28Speaker 3: Have five hundred thousand pictures of cats or five hundred
- 15:32Speaker 3: thousand images of whatever it is.
- 15:35Speaker 2: We've been asked for some strange things.
- 15:39Speaker 3: Now, we obviously individually are not going to go out
- 15:42Speaker 3: and find those items because it will take a very
- 15:44Speaker 3: long time. But what we'll do is we'll build an
- 15:47Speaker 3: understanding through AI of what a cat looks like in
- 15:50Speaker 3: this scenario. Then there will the AI will train on
- 15:53Speaker 3: this particular information. Once it's trained on the information is
- 15:56Speaker 3: able to gather that information, it will go and gather.
- 16:00Speaker 2: Huge inventory's images of cats.
- 16:05Speaker 3: Now, once that image then gets fed back through, there
- 16:08Speaker 3: will be certain criteria that the AI will then identify
- 16:12Speaker 3: within that information and what it's looking for in terms
- 16:15Speaker 3: of the annotation, which means that if certain items that
- 16:18Speaker 3: have been annotated are not in line with what the AI.
- 16:22Speaker 2: Is expecting effecially, they will be rejected.
- 16:25Speaker 3: If it's rejected effectively, that then will be rejected for
- 16:30Speaker 3: you know, one of two reasons is rejected because it's
- 16:33Speaker 3: not been fully done properly, but you can see that
- 16:36Speaker 3: they've made an effort, this is something we need to
- 16:38Speaker 3: kind of figure out, or it's been done, you know,
- 16:42Speaker 3: with no intent to kind of do it correctly, in
- 16:44Speaker 3: which case then penalty of should apply. So people shouldn't
- 16:47Speaker 3: be in fear of contributing data sets because they might
- 16:50Speaker 3: get it wrong as long as they kind of you know,
- 16:54Speaker 3: making efforts to do it correctly, but there should be
- 16:57Speaker 3: penalties in place to kind of ensure that people are
- 16:59Speaker 3: doing it properly.
- 17:00Speaker 1: Definitely. I saw that Perceptron also just launched a ten
- 17:05Speaker 1: million dollar AI data Fund to fund different teams and
- 17:11Speaker 1: startups for AI data. This is super interesting. How can
- 17:15Speaker 1: you explain the data fund and the goal behind it
- 17:20Speaker 1: and how it ties into the greater Perceptron network.
- 17:25Speaker 3: Absolutely so, I've been in crypto for a long time now.
- 17:31Speaker 3: Originally I was the founder of the House of Crypto,
- 17:34Speaker 3: which is a YouTube channel, and over kind of you know,
- 17:38Speaker 3: my time in crypto, I've watched a lot of different
- 17:41Speaker 3: projects come and go, and really what kind of like
- 17:45Speaker 3: led me towards building perceptual to identify that actually I
- 17:48Speaker 3: really like crypto and blockchain technology and what it can
- 17:51Speaker 3: really offer the world. But it's very rare to kind
- 17:54Speaker 3: of think it of a way that in which blockchain
- 17:56Speaker 3: is an essential component of a particular aspect and can
- 18:00Speaker 3: actually improve the way that a service or business works.
- 18:04Speaker 2: I think when it.
- 18:05Speaker 3: Comes to AI, you have a few different ways in
- 18:09Speaker 3: which blockchain technology is the quintessential technology to be able
- 18:14Speaker 3: to kind of push things forward, right.
- 18:19Speaker 2: And because of that, I.
- 18:21Speaker 3: Want to kind of showcase blockchain technology and what it's
- 18:25Speaker 3: able to achieve.
- 18:26Speaker 2: And by building.
- 18:30Speaker 3: This particular business, been able to kind of gather, you know,
- 18:35Speaker 3: to solve a real issue, shall we say, And what
- 18:37Speaker 3: better way to showcase the issue being solved by giving
- 18:41Speaker 3: the opportunity to a lot of crypto or sorry AI
- 18:44Speaker 3: projects to come and use our quality data basically from there.
- 18:51Speaker 3: So essentially right now, it's to one give shine the
- 18:56Speaker 3: light shall we say on block tin technology and why
- 19:01Speaker 3: you know it can be the future and actually decentralization
- 19:04Speaker 3: works far better than centralization in one way like this.
- 19:07Speaker 3: And then also to showcase, sorry, to enable AI projects
- 19:12Speaker 3: to build out into the future, because obviously they will
- 19:16Speaker 3: be future paying customers, et cetera. And yeah, to sort
- 19:21Speaker 3: of move forward, give something back to the community, et cetera.
- 19:25Speaker 1: That's very cool. What's the timing on that data fund?
- 19:29Speaker 1: And if there are teams that are looking to get
- 19:33Speaker 1: funding or you know, get their startup out there into
- 19:36Speaker 1: tie into perception, n what's the best way to get
- 19:39Speaker 1: started on that.
- 19:42Speaker 2: So basically right now we are open to taking.
- 19:48Speaker 3: On board work sorry, right now, we're open to onboarding
- 19:51Speaker 3: AI customers who are looking to kind of make the
- 19:55Speaker 3: most of that fund. Projects that are working in the
- 19:57Speaker 3: crypto industry, perhaps in Web two as well, who are
- 20:00Speaker 3: looking to gather data in order to use for building
- 20:04Speaker 3: up their models, et cetera. Those companies probably best if
- 20:07Speaker 3: they reach out through x our Twitter account. That Twitter
- 20:10Speaker 3: account can be found through the website, which is I
- 20:13Speaker 3: think you're probably going to link it down blow definitely
- 20:16Speaker 3: down blow here, maybe the Twitter page as well. So
- 20:18Speaker 3: best to kind of give us a message on Twitter,
- 20:21Speaker 3: and then from there we can connect you through the team,
- 20:23Speaker 3: discuss the kind of requirements and needs that you're looking for,
- 20:27Speaker 3: and then kind of kind of process through that.
- 20:31Speaker 2: Awesome.
- 20:31Speaker 1: I'd love to you know, we don't have a lot
- 20:34Speaker 1: of time left, but I'd love to hear a little
- 20:36Speaker 1: bit of the long term vision and then the short
- 20:38Speaker 1: term heading towards the token launch. So in the next
- 20:43Speaker 1: twelve to twenty four months, how do you hope and
- 20:46Speaker 1: expect that Perception will grow sort of be a fierce
- 20:50Speaker 1: competitor in AI data space.
- 20:54Speaker 2: Good questions.
- 20:55Speaker 3: So, our main kind of goal right now is to
- 21:00Speaker 3: build out this kind of data questing platform and also
- 21:03Speaker 3: to move more towards live business intelligence. Now, right now,
- 21:09Speaker 3: there are existing companies within crypto that are gathering data
- 21:13Speaker 3: and selling data on to other companies. Okay, so it
- 21:17Speaker 3: might be gathering huge numbers of tweets or images for
- 21:24Speaker 3: training text to image like mid journey type models or
- 21:28Speaker 3: video whatever it might be.
- 21:30Speaker 2: That's fine, that's cool.
- 21:31Speaker 3: But what I'm kind of particularly interested in where I
- 21:34Speaker 3: think decentralization and having huge numbers advantage points actually becomes
- 21:38Speaker 3: really valuable is in building business intelligence building business intelligence models. Now,
- 21:45Speaker 3: this is very useful because where you're looking at the
- 21:49Speaker 3: Internet from right now compared to where I'm looking at
- 21:51Speaker 3: the Internet from, we will see different stuff. You go
- 21:54Speaker 3: on Amazon, you go on Facebook, for example, meta if
- 21:58Speaker 3: you go on Facebook right now, they will show you
- 22:00Speaker 3: ads based on the credit card that you personally have,
- 22:04Speaker 3: based on how much money you make. That's meaning that
- 22:07Speaker 3: lots of people are seeing lots of different things, lots
- 22:09Speaker 3: of different information. That information is very valuable to people.
- 22:14Speaker 3: It's very valuable to e commerce companies who are selling stuff.
- 22:17Speaker 3: They want to know where their products are ranking on
- 22:20Speaker 3: Amazon today in New York compared to where it's ranking
- 22:22Speaker 3: on Amazon in Dubai versus Brazil, wherever it might be.
- 22:26Speaker 3: To gather this kind of data and information is something
- 22:28Speaker 3: that's very very difficult for a centralized model to be
- 22:31Speaker 3: able to see it through all these advantage points, and.
- 22:33Speaker 2: It's where our real kind of mote.
- 22:35Speaker 3: Our separation from the rest of the market comes in
- 22:37Speaker 3: as a decentralized model. Because it's easy to build the
- 22:41Speaker 3: tech to build out a node. The nodes that people
- 22:45Speaker 3: are downloading are not particularly difficult, But what is very
- 22:48Speaker 3: difficult is to build a network of people in the
- 22:51Speaker 3: number of hundreds of thousands. And I've learned this through
- 22:55Speaker 3: my YouTube channel building and following us through undred k
- 22:58Speaker 3: over You know, it took me five us CIX years
- 23:00Speaker 3: to be able to build that YouTube Channel's difficult thing
- 23:04Speaker 3: to do something that centralized companies are not able to do. So, Yeah,
- 23:08Speaker 3: building towards business intelligence products, stuff that can be used
- 23:11Speaker 3: for e commerce, comparison sites, quant trading in cryptos. Sentiments
- 23:17Speaker 3: such a big important thing. Gathering sentiment analysis, selling it
- 23:21Speaker 3: to crypto exchanges to trading deaths huge.
- 23:25Speaker 1: Yeah, that is huge level the playing field for people
- 23:31Speaker 1: to earn in AI and not just contribute all day
- 23:35Speaker 1: to the next version of Claude and and open AI
- 23:39Speaker 1: and being the product. So I'm wishing you can team
- 23:42Speaker 1: all the best on it. I know it's live right now.
- 23:45Speaker 1: I'm going to leave all the links to the site
- 23:48Speaker 1: and the socials and the apps in terms of heading
- 23:52Speaker 1: towards the main launch and eventually people converting their points
- 23:55Speaker 1: into tokens. Is there any tenative of timeline that you
- 24:01Speaker 1: that you're seeing. Is this something that's happening in the
- 24:03Speaker 1: near future.
- 24:07Speaker 3: Yeah, So we moving towards the kind of TG I
- 24:12Speaker 3: said to you before, like it's a very dangerous thing.
- 24:14Speaker 3: I know everybody everybody always wants to know when it's
- 24:17Speaker 3: the same in crypto, everybody always wants to know when
- 24:20Speaker 3: it's like we all really know the bigcoin right now
- 24:22Speaker 3: is going to go up at some point. Everybody wants
- 24:24Speaker 3: to know when is the best time to buy. It's
- 24:26Speaker 3: actually the most difficult question. We see a TG coming
- 24:30Speaker 3: very soon. It's difficult for me to know exact dates
- 24:33Speaker 3: because that hinges on exchanges, how long it takes for
- 24:37Speaker 3: them to do their due toilgences, et cetera. So you know,
- 24:39Speaker 3: I'm not going to promise anything because I'm waiting for them.
- 24:42Speaker 3: But we're looking at the next kind of eight eight weeks,
- 24:46Speaker 3: eight to.
- 24:46Speaker 2: Ten weeks to be generous. Okay, well that sounds great.
- 24:49Speaker 2: It could be much it could be much sooner than that,
- 24:51Speaker 2: but I'm just OK.
- 24:52Speaker 1: So, what you're saying is such factor with the points system,
- 24:58Speaker 1: check out the data set better earlier than later.
- 25:03Speaker 2: Yes, absolutely awesome.
- 25:05Speaker 1: Well, I'm definitely going to check it out. It sounds enticing.
- 25:08Speaker 1: I don't know how many cats I can take photos of,
- 25:10Speaker 1: but maybe there's some part of expertise of kind of
- 25:13Speaker 1: data AI information that applies to me to get extra points.
- 25:17Speaker 1: Otherwise I'll be going full passive on all my extra
- 25:20Speaker 1: phones and computers to earn while I'm working away on
- 25:24Speaker 1: the show. I appreciate your insights into this and making
- 25:28Speaker 1: AI more fair for everyday users, Peter. It's something that
- 25:33Speaker 1: we really need and tying in decentralized networks super important.
- 25:37Speaker 1: So wishing you and the team all the best moving
- 25:39Speaker 1: forward towards launch and in the next few years in
- 25:42Speaker 1: that grand vision that you gave. And thanks for taking
- 25:44Speaker 1: the time to come on the show.
- 25:47Speaker 3: Thank you very much for having me. Really really enjoyed
- 25:49Speaker 3: chatting about AI data. Look forward to look forward to
- 25:54Speaker 3: seeing seeing some future inter fees to come on the
- 25:56Speaker 3: show again.