The value and power of Artificial Intelligence is growing dramatically every year, and will soon dominate the internet – and the economy as a whole. However, AI tools today are fragmented by a closed development environment; most are developed by one company to perform one task, and there is no way to plug two tools together. SingularityNET aims to become the key protocol for networking AI and machine learning tools to form a coordinated Artificial General Intelligence.

SingularityNET is an open-source protocol and collection of smart contracts for a decentralized market of coordinated AI services. Within this framework, the benefits of AI become a global commons infrastructure for the benefit of all; anyone can access AI tech or become a stakeholder in its development. Anyone can add an AI/machine learning service to SingularityNET for use by the network, and receive network payment tokens in exchange.

SingularityNET has three long term goals.

The first goal is to be the hub of open AI technology.
The second goal is to accelerate the development of open AI, leading ultimately to a working artificial general intelligence (AGI).
The third goal is to promote projects and technologies agreed to be broadly beneficial in nature, to mankind and to AI and AGI progress.

SingularityNET’s economic logic is optimized for achieving those goals.

To become the hub of open AI, SingularityNET needs both a protocol for AI Agent coordination and an open market for buying and selling services. The protocol lets AI Agents work together, it makes them interoperable.

The function of all markets is to price goods efficiently, so our first objective is to reduce friction and costs in machine to machine transactions. SingularityNET is a decentralized, open network. It should not be dependent on any exterior economy and should not discriminate access to participate. Anyone will be able to use or provide AI services on the network. It is a free, permissionless, and open market, and so our second objective is to maximize open and international access to a world of developers.

Second, SingularityNET was born to accelerate AI development. It is as much an ideological vision as an economic one. The network must both regulate behavior in the network, and incentivize the creation of new AI nodes and more AI processing. So our third objective is to ensure the economic logic facilitates dramatic growth.

Finally, SingularityNET aims to provide resources to projects and technologies that are democratically approved as beneficial, both in the present as well as in a forthcoming global AI economy. To maximize the success of SingularityNET, the economic logic must optimize for all three objectives. For that reason, we are introducing the AGI (artificial general intelligence) token. The economic logic of SingularityNET revolves around this single token, which is an ERC20 token, built on Ethereum.

The AGI token is straightforwardly a utility token, with a core purpose of being used to buy and sell AI transactions carried out by AI software wrapped in SingularityNET Agents, so it has mainly consumptive use. This token can be acquired during the token generation event, and will also be distributed after the token generation event, to participants in the evolving economy. The release of AGI tokens after the initial token issuance event will occur via a schedule to be described below, which can be modulated via democratic governance.

In 2018, efforts will focus on infrastructure development and tool prototyping, particularly of tools pertinent to language processing, biomedical analytics, image processing, finance and robot dialogue. This is a fairly wide scope, aimed at exciting developers in multiple domains.

Then for 2019-2021, the infrastructure development will focus largely on scalability in various regards (e.g. massive distribution of the blockchain mechanisms and core AI algorithms, extension to internet-of-things), and the Foundation’s AI development will encompass three vertical areas:

Social and emotional robotics

We will create tools enabling SingularityNET to be used generically for affective robots in the home, office, school, nursing home and hospital etc. We will leverage the close relationship of SingularityNET with Hanson Robotics, and also reach out to a variety of other robotics firms. Here the focus is on creating a decentralized “AI mind cloud” from which numerous robots can draw intelligence, to which they can contribute knowledge, and which they can use to communicate.

Biomedical analytics

A core functionality here can serve as the basis of many valuable benefit projects. We will focus on enabling the easy application of diverse AI tools to a massive number of interrelated datasets stored on the cloud. Homomorphic encryption will be valuable here for ensuring confidentiality of aspects of datasets when needed. Numerous clinical and biological research applications have need of massive cross-dataset AI analysis, and there is a lack of publicly available, flexibly configurable tools that provide this.

Cybersecurity

As this is an area in which SingularityNET itself will have acute needs, as will other blockchain-based projects which will be natural customers for SingularityNET. Machine learning can be used to identify patterns in prior attacks so as to better predict and identify new ones; and the rapid development of the world of cyberattacks lends itself very naturally to a decentralized collaborative AI platform.

Of course in spheres as rapidly changing as AI and blockchain it is critical to be adaptive, and it is possible the Foundation will end up shifting foci based on data not available now, or based on strong requests from network participants.

Conclusion

SingularityNET is a complex framework attempting to confront three complex and critically important goals concurrently

• Creating the world’s best marketplace for AI services – aimed at every vertical market including software, robotics and IoT and other hardware – via leveraging decentralized and open mechanisms to support contributions from the broadest possible variety of contributors, and support utilization by the broadest possible variety of users 46

• Creating a breeding-ground for the evolution and emergence of AGI – that is, a framework in which various AI Agents, carrying out operations at varying levels of general intelligence, can network and combine together to create increasing levels of synergetic general intelligence

• Working toward the common good for all humans and other sentiences, via fostering an open and inclusive economy and culture, and via dedicating a percentage of resources in the network toward the application of AI for beneficial tasks that promote general human well-being and the emergence of compassionate and helpful AI

These goals are not independent and the SingularityNET design pursues them in an intertwined way, so that progress toward any of the goals assists with progress toward all of them. None of these are simple matters, and nor are the underlying AI and blockchain technologies at our disposal especially simple; and so the design that we have outlined here is characterized by complexity of numerous sorts. However, the complexity is mitigated by modularization and a multi-layered design, which make implementation and maintenance feasible in spite of the various subtleties and the multitude of algorithmic and architectural demands.

The realization of the SingularityNET vision will require not only effective technical execution, but the concurrent and coupled growth of a large user community and a large developer community. Network effects on both these sides have the potential to rapidly propel the SingularityNET to a leading role in the international AI scene and potentially in the world economy as a whole. Further, the way the design fosters emergence of “the whole is greater than the sum of the parts” intelligence from the interaction of multiple AI Agents, provides a new level of network effect beyond what exists with current social networks, internet information retrieval systems, etc. but realizing these network effects will require effective execution on the business side as well as in the technology sphere.