SingularityNET Phase Two: Massive Token Utilization Toward Decentralized Beneficial AGI

A Proposal for the AGI Token Holding Community

Ben Goertzel


I encourage you to read through the documents we’ve posted today on Phase I and Phase II — and to review the proposal that we’ve put out there for you to vote on during the period Feb. 3–7.

This is an unusual SingularityNET blog post and among the most significant ones I’ve written — it doesn’t just describe a new feature or algorithm or application, it describes an envisioned rebooting and re-energizing of the SingularityNET project as a whole, with a view toward dramatically accelerating progress toward the project’s goals.

While the post is written for anyone who has an interest in SingularityNET and in the goal of beneficial decentralized general intelligence, it’s especially intended for those folks who have displayed enough interest in the project to purchase AGI utility tokens.

The SingularityNET Phase Two initiative that I’m going to describe here is the most substantial proposal regarding project mechanics and direction to be made since late 2017 when we held the initial AGI token generation event.

In the years since 2017, my colleagues and I have learned a huge amount, as well as building a lot of software and doing a lot of deep R&D. At this stage, we’re definitely only partially progressed along the path we set out on 3.5 years ago — to build a decentralized platform suitable for the emergence of beneficial AGI. But we’ve gotten quite a lot done.

We have the first version of a real working decentralized AI platform out there now, and we’ve built quite a lot of valuable and/or cool AI tools designed to work with it — everything from consumer apps like the Song Splitter and Rejuve COVID app to tools for analyzing genomics data and financial time series. We’re in the middle of rebuilding the OpenCog AGI system into a new system, Hyperon, designed to scale up hugely in connection with the SingularityNET platform. We’re working with Hanson Robotics on nursing assistant robots whose minds will back-end on SNet platform… and the more general SingularityNET-based SophiaDAO platform for humanoid robots across all vertical markets. We’re experimenting with applying our AI to DeFI with SingularityDAO and life extension w/ Rejuve, to decentralized processing infrastructure with NuNet … to clinical trial analytics and fintech and enterprise avatars with Singularity Studio … to sustainability with Arizona State University … and a whole lot more.

Fig 1. The Song Splitter app is available for free on Google Play and allows anyone to split vocal tracks from a piece of music — the app leverages the SingularityNET platform on the back end.
Fig 2. Covid Signals from is a soon to be released premier mobile app that tracks your health and offers advice based upon your condition.

But as much as we’ve done, it has to be admitted that SingularityNET and decentralized AI as a whole currently constitute only a minimal fraction of the global AI ecosystem. If we want to truly achieve the project goals and make it so that democratic decentralized AI is a major player in world affairs and helps guide humanity’s path toward a positive Singularity, we need to be doing even better and moving even faster. It’s with this in mind that I’m now proposing the major SingularityNET Phase Two initiative aimed at providing SingularityNET with the power to move forward as rapidly as possible to the next level.

Phase Two weaves together a number of different themes and plans, each with their own subtleties; to make things as clear and simple as possible I’m going to divide it here into four aspects:

  1. Shift from Ethereum to Cardano (to achieve new network functionalities)
  2. SingularityNET Layer 2 Networks (for driving massive token utilization)
  3. Launching a Stream of New AGI-ADA Tokens, to Drive Trillions of AI API Calls
  4. Moving Further Toward Decentralized Governance

In the rest of this post, I’ll briefly discuss each of these separately, emphasizing as well how they all tie together into a holistic plan for creating massive-scale AGI token utilization and thus pushing the SingularityNET ecosystem toward its goal of beneficial emergent AGI.

More detail on all these aspects of Phase Two can be found in the Phase Two whitepaper.

Also relevant is the SingularityNET Phase One Operational and Financial Review, which gives important context regarding the particulars of SingularityNET Foundation’s evolution and doings during the period 2018–2020, including many details that have not been publicly discussed before in any organized way.

Finally and in some ways most importantly, there are aspects of the SingularityNET Phase Two plan (specifically the launching of new AGI-ADA tokens that would expand the total token supply beyond the current 1 billion AGI tokens) that will require a token-holder vote for enaction — the proposal that will be put to a vote during the period Feb 3–7 2021 is linked here.

From Ethereum to Cardano

The first big part of SingularityNET Phase Two is the shift from Ethereum to Cardano platform. The good old AGI ERC-20 token isn’t going anywhere, but we’re working with IOHK (the company behind Cardano) to make a new AGI token on the Cardano platform — AGI-ADA as we are currently informally calling it. When the porting work is done you’ll be able to swap back and forth between AGI ERC-20 and AGI-ADA — the two species of the token will be swappable one for one so they’ll have the same value.

We’re building a new version of the SingularityNET Marketplace on Cardano which is envisioned to have a large number of new features enabled by Cardano’s Plutus smart contract language. Among other things, we’re working on an AI-DSL — a new domain-specific language that lets AI agents on the network efficiently and formally describe their properties to each other. Integration of the AI-DSL will constitute a big step toward enabling multiple AI agents to automatically assemble themselves into meta-AIs on the platform. We’re also in discussion with Cardano’s security team about achieving efficient multiparty computing and homomorphic encryption for AI algorithms within SingularityNET agents, via leveraging cryptography code already existing within Cardano’s codebase. And a lot more.

Driving Massive Token Utilization with SingularityNET Layer 2

The next major aspect of SingularityNET Phase Two is what we’re calling SingularityNET Layer 2 — SL2. In short, this is a bunch of software and tokenomic tools intended to support building secondary agent networks on top of SingularityNET.

Some SL2 networks will have their own tokens, some won’t need tokens. An SL2 network with its own token will pay a small fee in AGI tokens from each of its transactions — a bit like gas on the Ethereum network (but far cheaper than Ethereum gas fees are now) or transaction fees on ADA network (which are much more reasonable and likely to decrease).

A few SingularityNET spinoff projects that you’ve heard a lot about before — like Rejuve and NuNet and SingularityDAO — are slated to be some of the first SL2 projects. As well as the Awakening.Health nursing assistant robots and the SophiaDAO benevolent robotics AI network. But these SL2 projects SingularityNET Foundation has been incubating or co-founding are just the way to get SL2 off to a running start — we aim to stimulate the creation of a variety of amazing third-party SL2 projects, building their own communities and software in their own vertical areas and driving AGI token utilization in the underlying SingularityNET network.

Sophia, the world’s most expressive humanoid robot by Hanson Robotics — who will be playing a substantial role in SingularityNET’s future via the rollout of the SophiaDAO as a SingularityNET Layer 2 Network.

If you look at the charts in the SingularityNET Phase Two whitepaper [link XX], you’ll see that we’ve been working on charting out the potential of the various SL2 projects we’ve been incubating in terms of driving utilization for the SingularityNET network and if you. If you carry out the extrapolation rationally it comes out quite dramatically — you conclude that within 4–5 years we could be looking at hundreds of billions of API calls every quarter, trillions every year…

The above graphs display platform call projections over the next 91 years for the SingularityNET Layer 2 networks: Rejuve, NuNet, Singularity DAO, Xccelerando Singularity Studio, SophiaDAO and Awakening Health.

We will, of course, continue aggressively encouraging individual developers to build on the SingularityNET platform, and we anticipate this will be a major driver of network utilization, but we’re also seeing that a major driver for network utilization is going to be considering SingularityNET as a meta-platform that hosts Layer 2 networks.

Launching a Stream of New AGI-ADA Tokens, to Drive Trillions of AI API Calls

The aspects of Phase Two I’ve discussed so far are things that are almost definitely going to happen. The next piece I’m going to discuss is something which I believe would be of extremely high value for growing the SingularityNET ecosystem and massively driving token utilization, but which is a sufficient deviation from earlier SingularityNET plans that it must be put it to a vote of the community of AGI token-holders. To enact this portion of Phase Two, we will need a majority of the voting tokens to say yes to the SingularityNET Phase Two Proposal during the voting period from Feb 3 to Feb 7.

What we’re proposing — with a goal of promoting the future of the project and the large-scale utilization of the token — is to do a major Treasury action in early 2021. Specifically, we propose to start a process of progressively releasing new AGI-ADA tokens, above and beyond the AGI-ADA tokens that’ll be produced by swapping existing AGI ERC20 tokens.

In 2017 we minted a billion AGI ERC20 tokens in a Token Generation Event which attracted considerable public attention; the batch of tokens initially offered for sale was purchased in record time, roughly one minute. What we propose to do now is to release a billion additional AGI-ADA tokens — but to stream them gradually, rather than release them all at once. We propose to release them month by month, over 91 years, in exponentially decreasing tranches of 1.5% per month.

Although Phase 2 will last 91 years, the graph above only shows up to 2045. The specific dispensation of these monthly tokens is outlined in the Proposal. As described there, half the tokens will go to the Foundation to support its work building the SingularityNET network and ecosystem. Smaller fractions will go into pools to support like staking and reputation rewards. Some will go to current AGI token holders as rewards for ongoing loyalty.

30% of the newly released tokens will go into what we’re calling the DEEP Funding pool — Decentralized Ecosystem project funding — which will use Cardano’s Catalyst platform to allow the community to propose projects building on the SNet platform and then vote on which ones get allocated tokens.

Just to make it clear: All of us at SingularityNET Foundation understand that in the crypto world-changing one’s original tokenomic plan and issuing new tokens is a highly serious thing to suggest — and it’s frankly not what we thought 2021 would hold for the project, back when we did the original Token Generation Event in 2017. To be quite blunt about it, though, back in 2017 we didn’t foresee the crypto winter or the coronavirus. To flourish in an environment as unpredictable as today’s crypto, AI and tech ecosystems, a project needs to be agile and adaptive and roll with the punches — and the SingularityNET Phase Two proposal is an aggressive initiative to do just that.

If we move forward with the Phase Two plan, 91 years from now there will be 2 billion instead of 1 billion tokens. However, my own strong assessment is that, in this case, the overall value of the network in 2025 is going to be far more than twice what it would be if we didn’t release the new AGI-ADA tokens.

The SL2 networks we’re already working on have a clear capability to lead to trillions of AI API calls every year — and what about the ones we don’t know about. What about the decentralized Facebook-killer some currently unknown young hacker is going to build on top of AGI-ADA?

We can create SL2 networks and improve the platform and create AGI-ADA tokens and carry out all the other aspects of SingularityNET Phase Two even without these billion new tokens — but my strong assessment is, if we have to do it without them, it’s going to be a longer and slower path. And that worries me because our Big Tech competition, these are Trillion dollar companies and they’re not standing still. I want to be able to move forward full speed ahead, as fast as possible, extending what we’ve done so far to move toward beneficial decentralized democratic AGI — and Phase Two as we’ve designed it is intended exactly for that.

Progressive Decentralization: Making It Real

As a final important point, I now want to talk about the “democratic and decentralized” part of the SingularityNET vision. The DEEP Funding pool mentioned above ties in closely with the final aspect of SingularityNET Phase Two I want to talk about here — which is moving a bit further toward decentralized governance.

SingularityNET was founded with a philosophy of progressive decentralization — gradually moving from the initial condition of Foundation-based governance of the network, toward a final condition in which the network is a fully decentralized DAO. As part of the Phase Two initiative, we aim to take a significant step in the direction of decentralization, while still for the time being retaining the efficiency of having an organized Foundation carrying out a significant fraction of the work of growing the network.

We’ve had a Supervisory Council elected by the token holders for quite some time, but in Phat Two we want to beef up their role — we want to give them more power and more work to do, and also start paying them something in compensation for this added responsibility.

Going forward into Phase Two the SingularityNET Foundation will commit to releasing an in-depth financial and operational report each year, and one of the jobs of the beefed-up Supervisory Council will be to review this for the token holding community. The Supervisory Council will also have the job of reviewing all executive hires for the Foundation on the behalf of the token holding community.

Looking back on the last 3.5 years of SingularityNET, I’m proud and excited about everything we’ve accomplished so far — and I’m even more excited about what we’re going to do in the next phase. I feel we are poised for a rapid exponential increase in token utilization, which will kindle a major leap toward democratic, decentralized AGI. My colleagues and I deeply value the cooperation of the SingularityNET community in setting the stage for the next phase of the network’s evolution.

And once more I encourage you to read through the documents we’ve posted today on Phase I and Phase II — and to review the proposal that we’ve put out there for you to vote on during the period Feb. 3–7.

Join Us

We hope you will join us in our mission to democratize AI and to use its transformative powers to enact real, positive and lasting change. The best way forward is to come together and work practically towards creating a better future.

If you are looking to monetize your AI services or create new ones, we invite you to learn more about the nature of our platform and what it has to offer by visiting the SingularityNET developer portal.

We are proud of our developers and researchers that are actively publishing their research for the benefit of the community; you can read the research here. For any additional information, please refer to our new roadmap. To stay informed about all of our developments, please subscribe to our newsletter. You can connect with our passionate and talented community by visiting our community forum.