SingularityNET AGI on Cardano ADA Progress | March Insider Update 2021

ben gertzel here i want to talk to you a little bit about what's been happening with singularitynet for the last month and you know what we have to look look forward to during the next few months which is is really really exciting so february 2021 i mean this has been a legendary month for for singularity in a whole bunch of ways i mean that the vote that we had in early february approving the phase two initiative i think this is sort of the biggest threshold event in the history of singularity in that since the original token generation event in the in in late 2017 i mean between the the collaboration with cardano the agreement to release a billion new agi tokens on cardano platform you know month by month uh starting pretty soon and the you know reorganization of governance to move further toward uh progressive decentralization i mean we're really we're really setting things up to enable tremendous growth in in the next months and the next few years i mean growth in the intelligence of the ai and the network growth and the utilization of the network and growth and in the economic value of the network and you know the economic return and the humanitarian return to be received by by everybody participating in in in the network so i think uh incredibly incredibly excited about you know the support we had from the community in voting for the phase two initiative by such a large margin and most of all about about what this is going to enable going forward so you know you're going to hear a lot more about that in in the next few weeks and and months i've been working a lot with the cardano team on on technical aspects in terms of working toward getting the agi token actually running on the cardona network getting getting the marketplace there and then some some deeper aspects like the ai dsl ai domain specific language that you know once we have the marketplace and the protocol fully poured the cardano this will allow the ai agent during cardano platform to communicate with each other in a richer and deeper way so the agents can understand what each other are doing and why what properties they have so that ai agents on the platform can coordinate and and cooperate together in a you know a richer and more emergently intelligent way than is possible now on the ethereum platform i've also been working a lot with the cardano catalyst team and a bunch of our internal singularity net team members and the singular supervisory council on the the deep funding pool the decentralized ecosystem projects pool which is a another key part of phase two and i mean we're i think we're gearing up to make this one of the most uh exciting tech project funding you know incubator accelerator plus plus programs on on the face of the planet i mean we're looking at using this to put funds into community projects you know enhancing the singularity net infrastructure to projects building building dapps and various applications on top of singularitynet so it's to get more and more utilization of the platform on more api calls and then you know researchers developing amazing new ai tools core ai agents to run on singular unit that can that can then be be accessed by products uh serving serving customers running on the platform so yeah between infrastructure dapps utilizing the platform and core ai tools i mean all of these things fertilized by the by the deep funding i think and this can go a long way toward bringing the singularity net network the massive token utilization that it needs to really grow both commercially and grow toward emerging emergent general intelligence and that you know i can't go into detail here on all the things we're working on to make phase two a reality but there's a lot it all it all connects together and it's just just an incredibly exciting time so i want to i'll answer a few questions from the community now to uh sort of uh flesh out a few of the details and the way that's directed by what people in the community are are interested in let's let's see question uh where can i see a roadmap for agi on cardano well the phase two white paper the phase two proposal at the end you know gave a robot for the next several years but the the other on there for 2021 are the outline of a of a roadmap for uh for singularitynet project in in the near future but we are working on a more in-depth roadmap uh sort of uh in in that direction there's going to be no big surprises there but there'll be there'll be a bit more detail about sequencing and and timing so yeah you can you can expect that pretty soon but if you want the high level it is there at the end of the the phase two white paper a question how will singularity net scale let's say a million people are sending requests to one ai agent how it will be able to deliver to all of them well you know scaling is a buzzword that boils down to a lot of different things when you when you look uh you know under the hood so i think the cardinal integration will be a a significant part of it cardano just fundamentally has a more scalable architecture than ethereum you know lower cost and and higher speed i mean we already have algorithms allowing singularitynet to scale in terms of peer-to-peer search for ai agents meeting certain properties and so forth it's just the the practical atomic operations in these algorithms on ethereum are are are very slow and that will be solved with the move to to cardona no you know there's a lot of other aspects to scaling that that happen sort of once a customer has reached a certain ai agent i mean that yeah that ai agent may need to make multiple copies of itself to serve multiple customers you know may need to make multiple copies of the the data that it's accessing if a lot of its customers are having up all that data at the same time and i mean this uh this is part of what we're referring to as infrastructure as a service which is also written about in the in the phase two white paper i mean this frankly is multiple years of work to get scalability in singularity as a decentralized infrastructure up to the level of what you have in like aws or or azure or something i mean we'll be getting there getting their bit by bit but i mean these are trillion dollar companies who have put a tremendous amount of work into into scaling and that you know we're we're doing it step by step and then another interesting aspect here is the new net project which is a a spin-off of singularity net which we expect to pick up more and more momentum as 2021 continues including very likely launching its its own token perhaps as a cardano native asset but nunet allows you know processing power on mobile devices internet of things devices just uh cpus and in laptops and desktops allows all these compute resources to be to be pulled in to the benefit of ai agents running on decentralized networks you know initially singularity and also other networks that may connect to singularitynet so i mean i think that nuna in a way lets us go even beyond the infrastructure as a service of big tech companies offer and you know cardano can be in in a way that the glue that lets us pull these infrastructure as a service tools both traditional ones and and noona pull them together to provide you know additional types of scalability on the back end of of singularity ai agents so yeah it's a we'll see a big boost in scalability with the point to cardano but scalability has a lot of dimensions and it's going to be increasing step by step question neuralink and singularitynet is this a match made in heaven well i mean ai and brain computer interfacing are a match made in heaven right i mean if we can if we can plug the ai into our brain i mean then you know a as humans our own thought process is going to be enhanced by what the ai can do i would love to be able to access you know search engines uh calculators and and uh ai systems of all sorts at the speed of of thought right but additionally the ai the ai is going to be able to learn from human thinking and and and the and the human mind and you know really putting brain computer interfacing and ai together is pretty much you know that's going to be the best way for ai's to really learn what it what it is to be human on a profound level i mean we can we can go a long way with robots like sophia which we're pursuing you know heavily through the sofia collective sophia dial collaboration with hanson robotics and through the the grace humanoid robot for elder care that we're developing in another joint venture with hanson robotics called awakening health so these you know these humanoid robots can look can look you in the face and they can they can empathize with you and that that's one way to get human values culture and understanding into into the into the robot but of course brain brain computer interfacing i mean this is a this this adds a whole other dimension i mean neuralink is very cool company i mean that's it is one among many brain computer interfacing initiatives on the on the on the planet now and you know partnering with them or any other of the many serious brain computer interfacing initiatives out there certainly makes a heck of a lot of sense i i would say i i have a bias towards sort of open source hardware and open source software i mean this especially where you're plugging stuff into your brain you'd really like to be able to to audit the code and be be certain that like you're not going to get you know adware in your neocortex or something so i would i would have a bias toward partnering with brain computer interfacing projects that are you know as open and transparent and and decentralized as as possible and you know this should be a really interesting area to explore going forward question don't we really need a consciousness facilitative ai if we raise human consciousness i'm sure this would have a profound impact on raising the levels of a consciously aware artificial general intelligence would result in a benevolent ai singularity well that's i mean obviously i selected that question because yes indeed i do think we need a human consciousness facilitating agi i mean i i think the idea that we're just going to create a superhuman agi mind you know in in a box somewhere in in an isolated computer network somewhere that's going to be not only super intelligent but so benevolent so understanding that we'll solve all humanities problems i mean that's it's not totally impossible but it's probably not the way things are going to go i mean there's a big risk that isolated the ai system is not going to understand what humans really want or need and it's going to go in some other radically different direction and i mean i think by far the best way to get to an agi that's really able to help humanity resolve all all its uh issues and move on to a profoundly better state state of existence i mean the best way to do that is for humans to be uplifting their own consciousness understanding themselves better you know loving each other better help helping each other better working toward a better and better society and a better way of being human and for the for the agi to grow up as as part of this consciousness expansion on on the part of humanity and i mean if if we can do that this is really how you get to a beneficial agi now of course i'm you know i'm not completely oblivious to what's going on in the world now for real i mean i lived in hong kong for nine years i saw all the political struggles there a lot of stuff very very disturbing i mean i've uh relocated back to the us as my main base of operations uh as of last year although i'm gonna once covered travel restrictions are gone i'm gonna be roaming far far and far and wide again for sure and i'm in us uh you know we had we had donald trump we we have a disturbing percentage of uh u.s population believing q and on right i mean we we've got a lot of screwy things going on in humanity and human systems all over the planet but i think if if among all this mess we can get serious human consciousness expansion to happen so that you know there's a process and dynamic of a larger and larger percentage of the human population awakening themselves loving better understanding themselves and and other others better opening their minds to you know the the amazing possibilities that the future holds you know if this becomes more and more of a thing more more of an aspect of humanity we still have a chance to raise up our agis to raise them up so that their own growth and expansion of their consciousness is like synced up with the positive expansion of human consciousness and for this to be you know the dominant dynamic in in the emerging agi mind and of of course you know with practical work on singularitynet um we're often focusing on much more nitty-gritty stuff like how do you get more ai developers to put stuff on our platform how do you get how do you get more more api calls into the platform how how do we increase you know the the economic magnitude of the of the of the network and we're we're building infrastructure we're dealing with a lot of nitty-gritty stuff but i mean in the end what we're after with singularitynet is the same as as it was in 2017 when when we launched a project right it's it's democratic decentralized benevolent artificial general intelligence i mean so it's it's assembling practical ai systems that solve real problems for real people in the real world now but in a decentralized way and on a platform where the multiple ai's in the platform can easily cooperate and combine with each other to get to move to higher and higher levels of emergent general intelligence and i mean i think if if you can get to agi that way coming out of the cooperative activity of a diverse pool of agents created by a diverse pool of people to serve the needs of a diverse pool of customers i think and and if you know a reasonable portion of these agents and the people using these agents are are are themselves moving in the direction of you know expanded more benevolent consciousness i mean i think i think we have at least a decent chance of creating a really benevolent agi that's raising its own consciousness up as it becomes more intelligent and can then can then help human beings in in really profound ways question all right coming back to the practical here once nunet launches an increases in number of validator nodes could agi switch over from operating on ethereum and cardano to operating on newnet well known that newnet isn't a blockchain right i mean nunet is really a general purpose decentralized computing fabric which could could leverage multiple blockchains multiple kinds of hardware devices and conserve multiple decentralized computing networks so i mean you could run singularity net nodes on on nun you could run ocean protocol nodes on on i mean so it's nuna is a pretty general sort of decentralized computing fabric not not everything needs to be a blockchain i know we we hear all the time in singularitynet community why doesn't singularity make its own blockchain i mean i think sort of because of the history of the blockchain space there's the idea that every good project should form its own blockchain but i mean it doesn't have to be that way any more than like a google had to make it it's its own os kernel i mean i mean you can linux kernel is good you can use the linux kernel you can build a lot of amazing stuff on the linux kernel without needing to make your own kernel and cardano is is good in the same way the linux kernel is good you can put a lot of stuff on top of cardano without needing to build your own your own cardano and the blockchain is a super important part of the stack if you want to be making decentralized computing or decentralized a ai networks but you know there's a lot of other parts to the stack so i mean we're singularitynet is living just above cardano in the snap stack then you know networks like rejuvenate and so forth we can think of as layer two networks that that are living living above singularitynet you can think of things like opencog hyperon or singularity studio with its enter for enterprise software products is even like layer three they can leverage all the different layer two networks on top of singularity net something like uh toda ip protocol you can even view as living below cardano below below the blockchain in the stack that's like tcp on on decentralizing encryption or something like an even lower level protocol that canalo could make use of so there's there's a lot of layers in the stack and it's good that uh cardano is an amazing blockchain singularitynet is already a pretty solid solid decentralized ai network and nunet's position is really on top of that as a layer 2 network for singular and other decentralized networks and that's it that's great i mean you know the centralized tech companies that the googles microsoft tencent to the world and so on they they've got very rich multi-layered tech stack already and we're really building out all those layers uh at the same time as fast and as well as we can in the decentralized ecosystem and it's a you know the big talk big tech companies are very worthy competitors on the other hand i would say we we also have some real advantages in the decentralized ecosystem and that actually that ties into another question so what what do you think of this situation with the timnit jeberu and uh and google's ai ethics uh brouhaha so yeah i mean i i felt felt that somewhat personally just because uh tim that is is ethiopian and you know singularitynet has had the office in ethiopia since the beginning i've been doing ai and robots work in ethiopia since 2014. and um you know there's amazing it's a big country there's almost 100 100 million people there's an amazing amazing talent there and it's uh it's great to see ethiopians and ethiopian women in particular going out and pervading every aspect of the modern tech community and then you know it's uh it's not at all surprising to me that uh some of the more you know dazzling female ai minds from the developing world may run into sort of uh cultural issues and and problems with the attitudes of people in in in big tech companies i i mean in the end google start out pretty cool as i mean i've visited google office first in the year 2000 it was it was it was it was great it was like a bunch of uh grade a geeks just solving hard computer science problems and rolling them out to the world amazing amazing stuff right but you know google has now just become in many ways like any other any other big company right so when you look at the processes that they want people to go through to get papers published in within google which was wrapped up with the the issues that google had with its ai ethics group i mean if you're working at microsoft or working at ibm or you know ubs there's all sorts of processes to go through to to publish papers that's how the corporate world works they they want to maximize shareholder value and they want to control what their staff are doing and if you're a researcher there you know you're not a academic professor i mean you're you're serving the man you're you're serving the company's needs and i think this mess with google's ai ethics group has just reinforced to the world that you know these big tech companies they're not that different they're not that special that i mean you know google removed don't be evil as its corporate model a long a long time ago right i mean these are big companies they're out to maximize shareholder value and you know they're not necessarily less ethical than other big companies out there they're not necessarily more ethical than other big companies out there either i mean i i do think uh to me the biggest eye opener with this whole tinder gebru thing was just how incompetent google wasn't managing all this i mean i i'd sort of as a as a die-hard cynic about big companies i i sort of thought well google was going to hire these ai ethics people to publish papers that said nice intelligent sensible things about ai ethics and that was a sort of disinformation campaign well well they're going to keep on doing the same old unethical things to to make money and maximize shareholder value right and so it seems like not everyone inside google was maximally aligned to to play that game right and so that's uh that's a little a little strange and confusing but uh i mean i'm i'm sure i'm sure they will write the course and and play their uh pr disinformation campaigns better in in the future on the other hand i do think there's some some long-term damage done and i think not just google with all the big tech companies like they're no longer they're no longer perceived as the the cool place to work for for young ai developers or young computer hackers i mean by now going to work at google or facebook is a bit more like going to work in a wall street bank or something like you do it because it pays a lot of money and it's good for your for your resume but you don't do it because you think you're like saving the world or because it's the most exciting or interesting thing to do and i think i think that's very positive for decentralized projects like like singularity net right because if if you know very talented ambitious young developers and ai blockchain in other fields i mean if they if they don't want to jump into these big tech companies because the big tech companies are no longer the cool thing i mean it's more likely they're going to throw in with you know crazy decentralized projects like like singularity net or or cardano so i think as as a big tech company's cool value fades you know that we we may be able to pull more and more you know high talented people into into the decentralized ecosystem which brings us to the next question now what we do to attract more developers in the singularity ecosystem well yeah i think you know the port to cardano at least will get rid of a couple reasons not to put your code in singular that platform like high cost of gas and slow speed of ethereum transactions so that that's that that certainly helps but that that just removes some bad things it doesn't add good things right i mean bottom line is as a as a protocol a platform or a meta platform i mean singularity that needs you know needs a demand side in the supply side the more demand there is for ai services on the platform the more motivation there is for people to put their ai into the network providing supply and and vice versa the more ai has been put on there the more supplies there then the more people will want to want to use the network the more demand there will be so you know we've there's a two-sided network effect that needs to be seated which to be honest we haven't made that much traction on so far we've built and launched some amazing software but we haven't really gone far enough to get utilization of the of that of that software and i think you know we've got a number of strategies for that going forward i think it's the time for this to happen so the the deep funding pool that i mentioned which is part of the the phase two initiative this can be a really big part of it but we don't you know we don't just want to be paying people to put stuff on on the platform and that's not sustainable so what what we need to do is use deep funding to seed projects just to give them a little head start to get them started well they will then keep on using the platform ongoingly you know because it's good for them because it's providing with technical support they wouldn't have had otherwise or because because they're making money there as an augmentation to what they can get on on other channels but at least the the deep funding pool which will be funded by a steady stream of of agi on cardano tokens i mean at least the deep funding pool provides some economic firepower that if if used judiciously can can stimulate putting both core ai algorithms infrastructure improvements and and that's leveraging the network so all these things onto the singularity net platform i think some of the layer two networks that we've been building within singularity net foundation are going to be spinning out like a rejuve for life extension biology and the new new net i've our dimension for decentralized processing accelerando for decentralized media i mean i think all all these all these things through the logic of their own networks are going to create demand because they need ai services from singularity net network and by by juicing the supply side with these layer two networks gonna launch on top of singular unit that can conduce demand demand uh uh of a fair bit really i mean i think we also need to do the hard work of building out more infrastructure as a service tools underlying singularity net platform in order to really offer to developers as much as as the centralized platforms offer and that you know that will happen step by step meaning that that you know some some developers will be able to leverage singularity net as well as they could leverage centralized platforms this year in 2021 i mean some could right now and for some will need you know service fleet management or automated replication and fallback we'll need more of these infrastructure as a service functions to be there for for some product developers to really jump full-on into singularitynet so it is going to be a gradual process i mean i think deriving from our integration with cardano you know tools for multi-product computing and homomorphic encryption integrated with machine learning and reasoning algorithms this can can can help also because if developers want to do things in a you know distributed and secure way there's not a lot of platforms that that support that and the integration with cardano can let us let us get there faster so there's a bunch of different things there's there's no like one magic solution but i i i think all in all between deep funding the layer two networks and the move to cardano and building more infrastructure as a service i think we've we've got a lot of things that are going to pull more and more developers into the platform and then what you need you need to reach a critical mass and then then there's enough there that it just it just grows uh grows of its own momentum one more question so does singularity dao benefit the singularity net landscape so yeah singularly dao is a is a spin-off a layer two network of singularity net that uses singularity ai for decentralized finance and i mean it uses it specifically to build a sort of uh trading and then yield farming infrastructure on on top of lower liquidity altcoins so i mean you can you can bundle together a basket of of all coins into a single token called a dyna set which is a bit like a bit like an etf but the components are dynamically you know reweighted by ai ai running on singularitynet and then then you can also stake these dyna sets these etf like tokens you can stake these data sets and then earn yield tokens for that and there's there's hedging and futures based aspects there and i think you know first of all this benefits it benefits the singular network by using ai running on singularitynet right so that's that's token utilization right i mean there also will be an airdrop of singularity dao governance tokens to agi token holders and the particulars of that are are still being worked out additionally when we launch new spin-offs of singularity like say a nuna or rajuv i mean we can leverage singularity dao to help create liquidity and upgrade healthy markets for those tokens along with the other tokens bundled in the singularity dial down the sets right so i mean i think there's there's a lot of ways that singularity dial is going to drive uh drive single argument forward uh okay one more question on my list one with the the sofia dial launch well so i've been working closely with david hansen uh michelle singh and others on on what we've now been calling the sofia collective which is is sort of a collaboration of singularitynet and hanson robotics and and a few other partners which is is is aimed at uh you know putting decentralized open source ai behind sophia and other similar robots and avatars with with the future working toward a beneficial singularity right so you know we were talking earlier i was talking earlier about uh ai that facilitates expansion of human consciousness and that was really david hansen's goal with launching the sofia robot in in the first place you know hanson robots is putting a bunch of their ip so open source and and and to be managed by singularity dao and singularity net and singularity studio are our commercial spin-off for enterprise software we're going to take a bunch of the software that we're putting behind the the grace humanoid robot which we're launching with awakening health to help elderly people and elder care facilities we're going to take a bunch of that software and also open it up and put it put it out to the public through through the the sofia dow sophia collective initiative so that the goal there is to really build an open source community around you know creating benevolent generally intelligent humanoid robots with a great amount of human empathy and with m with their minds running in the back end on singularity net decentralized network and you know this uh there's a bunch of other aspects to this that you're going to hear about in in in the coming weeks so when we're focusing heavily there on visual arts and and music and video and sort of on the the aesthetic aspect of of ai and the ability of that to reach out to more people and pulling a broader community because i mean the idea with the sofia dao sofia collective i mean it's still at a formative stage in some ways but the idea is to pull in a whole bunch of people who may not be programmers or or technical minded and and pull them into helping collaboratively sort of shape the consciousness of sofia and her uh her sister and brother robots you know with with a view toward sort of shaping the overall you know benevolent emergence of the the global brain of humans computers ais ai's robots and and so forth but yeah i'm i'm getting very airy fairy here but i mean it's it's a very very cool exciting project uh there's gonna be some really interesting announcements about that before too long and i mean those who have been following singularitynet for a while you remember the sofia robot was like with me all the time when i was out there talking about singularity in 2017 we were doing a bit less with sophia for the last few years but but behind the scenes we were working on the ar software for natural language dialogue and humanoid robot control right so i mean now now that work in the background on that software i mean now this is a this is coming to fruition both in the grace humanoid robotics project which is applying the singularity based you know ai robot control software to elder care and in the sofia dial sophia collective project which is which is applying it more generally with the sort of arts and the aesthetics and ethics spin but uh yeah i think that's uh i think that's enough for now so yeah as i said this has been a really exciting month i mean getting the go-ahead from the community to proceed with the phase two initiative is just it's just uh incredible and uh you know we've got so many exciting things in in the work now this is going to be the most amazing year in the history of singularity and i think i'm going to be able to say that every year for some years to come i mean that's the nature nature of exponential growth and exponential change and i think we're entering a period of very very rapid exponential growth for singularity and decentralized you know democratic beneficial ai and agi [Music]

Channel: SingularityNET

Views: 10,303

Keywords: singularitynet, cardano, agi on ada, dr ben goertzel, blockchain artificial intelligence, ai cryptocurrency, artificial general intelligence, ai domain specific language, the technological singularity, charles hoskinson

Id: FSkNokF0nZs

Length: 36min 8sec (2168 seconds)

Published: Tue Mar 16 2021