PlatON 2.0 White Paper: Decentralized Privacy-Preserving AI Network | Part 1

Abstract ② [ *aY ]

In just the past 10 years society has witnessed the transition of analog to digital, and we are fast forward to a fully digital life. However, the data utilization rate is very low due to the high centralization of artificial intelligence, coupled with data abuse and privacy leakage. The value of data urgently needs to be intellectualized in order to be deposited and utilized, hence the growing call for a next-generation intelligent network known as web 3.0. PlatON brings blockchain, AI and privacy-preserving computation together to create a decentralized collaborative privacy-preserving AI network that takes data utilization to a new level. This network also serves as an infrastructure for autonomous AI agents and their collaboration that can facilitate the emergence of advanced AI and explore the path to artificial general intelligence. Based on a underlying blockchain network, we first establish a decentralized privacy-preserving computation network that connects data, algorithms, and computing power through privacy-preserving computation protocols. The developers can obtain the required resources at low cost, train A models and publish them to the network, where AI services or agents interact with each other to form a self-organized, collaborative AI network. Anyone can access AI technologies or become a stakeholder in its development, thus democratizing AI. The PlatON network create a new AI fabric that delivers superior practical AI functionality today while moving toward the fulfillment of PlatON artificial general intelligence visions.

Intelligence and privacy security in the digital age

According to Statista analysis[1], the number of connected devices worldwide is expected to reach 30.9 billion by 2025. Connected devices and services create enormous amounts of data, and IDC forecasts that by 2025 the global data will grow to 163 zettabytes (that is a trillion gigabytes). That’s ten times the 16.1ZB of data generated in 2016. All this data will unlock unique user experiences and a new world of business opportunities. Where once data primarily drove successful business operations, today it is a vital element in the smooth operation of all aspects of daily life for consumers, governments, and businesses alike. In just the past 10 years society has witnessed the transition of analog to digital. What the next decade will bring using the power of data is virtually limitless.

In Data Age 2025[2], IDC identified artificial intelligence and security as critical foundations.

Artificial intelligence that change the landscape. The emerging flood of data enables artificial intelligence technologies to turn data analysis from an uncommon and retrospective practice into a proactive driver of strategic decision and action. Artificial intelligence can greatly step up the frequency, flexibility, and immediacy of data analysis.
Security as a critical foundation. All this data from new sources open up new vulnerabilities to private and sensitive information. There is a significant gap between the amount of data being produced today that requires security and the amount of data that is actually being security protected, and this gap will widen. By 2025, almost 90% of the global data will require certain level of security, but less than half will be security protected.

The coming “intelligent web”

From WEB 2.0 to WEB X.0

Web 2.0, coined as such by O’Reilly and others between 1999 and 2004, is a platform that reached millions of users and facilitated communication, organization and collaboration. Today, more than a decade later, serious questions are being asked about the centralization, privacy concerns and security of the current Web.

Centralization: Web 2.0 has evolved to a point where large technology and social media companies dominate the market and hold vast amounts of personal data on users.
Privacy and security: With increasing amount of data being captured large data centers act as honeypots for organized crime.
Scalability: With larger data sets from billions of connected devices, there will be increasing pressure on existing infrastructure. Today’s client server model works well, but is not likely to scale for the next generation web.

The next generation of the Web was first named “Web 3.0” by John Markoff of the New York Times, but there is no definitive definition of Web 3.0. The Semantic Web, as proposed by Berners-Lee, is often used as a synonym for “Web 3.0,” which would process content in a human-like manner. Broadly speaking, the characteristics most commonly associated with Web 3.0 include the following.

Ubiquitous Connectivity, connect anyone, anywhere, anytime to anything that is open, trustless, and permissionless.
The Semantic Web, Web 3.0 will use efficient machine learning algorithms to connect data from individuals, companies and machines in a cryptographic way, and machines will be able to understand and intelligently process the data in a human-like manner.
The Intelligent Web, Web 3.0 is an evolutionary path to artificial general intelligence that can run Intelligent applications, such as natural language processing, machine learning, machine reasoning, and autonomous agents.
Self Sovereignty, Everyone is in control of his or her own identity and data. No need to rely on third parties, individuals can sell or exchange their data without losing ownership and privacy.

Nova Spivack[3] proposes that in the coming “intelligent web”, Web services are connected to autonomous intelligent agents, roaming the network and able to interact with one another, Web sites, and even people. As the intelligence with which such processes unfolds, in a totally decentralized and grassroots manner, vast systems of “hybrid intelligence” (humans + intelligent software) will form, which is the Metaverse. The network becomes increasingly autonomous and self-organizing, as structures that provide virtual higher-order cognition and self-awareness to the network emerge, connect to one another, and gain sophistication, the Global Brain will self-organize into a Global Mind.

Underlying technologies of intelligent web

Where Web 2.0 was driven by the advent of mobile, social and cloud, The intelligent web vision is built upon three new layers of technological innovation: blockchain, artificial intelligence and the Internet of Things. Its ubiquitous nature is underpinned by the growth of the Internet of Things. Artificial intelligence will be a crucial tool used to tag Web content, and a truly semantic Web will enable AI systems to leverage it in new and novel ways. Distributed ledger technologies such as blockchain will underpin an intelligently connected Web 3.0, by facilitating data exchange and transactions between divergent systems, manufacturers and devices.

The key to the leap from Web 2.0 to intelligent web remains the protection of data privacy and the ownership of data should be able to be controlled by individuals on their own. privacy-preserving computation is an emerging solution and technology trend, and is listed in Gartner’s 9 key strategic technology trends for 2021 [4]. privacy-preserving computation makes personal data more secure and private, allowing users to truly take ownership of their data, fundamentally balancing the contradiction between data security and data value, and completely resolving the safe and free flow of data.

Developments and Issues in Artificial Intelligence

Trends in Artificial Intelligence

During the last 5 to 10 years, the rapid growth of the Internet, mobile Internet and Internet of Things has generated enormous amounts of data. The increase in chip processing power, the popularity of cloud services and the decline in hardware prices have led to a significant increase in computing power. The broad industry and solution market has enabled the rapid development of AI technology. AI has been everywhere in human’s daily life, and AI has been applied in many industry verticals such as medical, health, finance, education, and security.

A growing number of governments and corporate organizations worldwide are gradually recognizing the economic and strategic importance of AI and are dabbling in AI from national strategies and business activities. A study by PwC on the economic impact of AI[5] on the world economy by 2030 reports that the emergence of AI will bring an additional 14% boost to global GDP by 2030, equivalent to a growth of $15.7 trillion, more than the current GDP of China and India combined. The global AI market will experience phenomenal growth in the coming years. In its 2019 Global AI Development White Paper [6], Deloitte projects that we forecast the world AI market to exceed $6 trillion in the next 2025, growing at a CAGR of 30% from 2017–2025.

In the field of mainstream artificial intelligence, Deep Learning has made breakthroughs in recent years, rekindling hopes for “human-like” AI. Claims that the “Turing test has been surpassed” and AlphaGo’s victory in the human-computer Go match have made the discussion of artificial general intelligence(AGI) a hot topic in the industry. Technology giants such as Apple, Amazon, Alphabet, Microsoft, and Facebook have invested heavily in AGI research and development, with Google spending $540 million to acquire DeepMind in 2014, Microsoft investing $1 billion in OpenAI in 2019, and according to a report on general AI by Seattle research firm Mind Commerce [7], investments related to general AI will reach $50 billion by 2023.

According to Mind Commerce’s AGI report, the global market for general AI for enterprise applications and solutions will reach $3.83 billion by 2025, and the global market for AGI-enabled big data and predictive analytics will reach $1.18 billion. By 2027, 70% of enterprise and industrial organizations will deploy AI-embedded intelligent machines, more than 8% of global economic activity will be done autonomously by some kind of AI solution, compared to less than 1% today, and more than 35% of enterprise value will be directly or indirectly attributable to AGI solutions.

Challenges with Artificial Intelligence

Data Privacy and Security Regulations

Machine learning technology, mainly deep learning, cannot be learned and inferred without enormous amounts of data, so enormous amounts of data becomes one of the most important resources for the development of frontier technology of artificial intelligence. Technology giants, especially those in China and the United States, have accumulated huge amounts of data through the Internet services, and as the value of data becomes increasingly prominent in the era of AI, these data will gradually evolve into an important asset and competitiveness of enterprises. According to IDC estimates, the global data volume is expected to reach 44 ZB in 2020, and China’s data volume will account for 18% of the global data volume, reaching 8060 EB (equal to 7.9 ZB) in 2020.

The more “intelligent” artificial intelligence is, the more personal information data needs to be acquired, stored and analyzed, which will inevitably involve the important ethical issue of personal privacy protection. Today, all kinds of data and information are collected all the time and everywhere, almost everyone is placed in the digital space, personal privacy is very easy to be stored, copied and spread in the form of data, such as personal identity information data, network behavior trajectory data, as well as data processing and analysis of preference information, prediction information, etc. It is foreseeable that in the near future, more and more artificial intelligence products will come into thousands of households, which will bring convenience to people’s lives while also easily accessing more data information about personal privacy.

Entering the 21st century, many companies worldwide, including Internet giants, have been exposed to data leakage and abuse. Google, Amazon, Facebook, Apple and other U.S. Internet companies have been fined by the EU one after another in Europe in the past two years for data privacy, monopoly, taxation and other issues, which has caused widespread concern worldwide and made people gradually aware of the importance of personal privacy protection. Countries around the world have successively introduced bills to further regulate the market. The promulgation of the Law of the People’s Republic of China on Network Security and the National Strategy for Cyberspace Security in China, and the General Data Protection Regulation (GDPR) in the EU have had a profound impact on the protection and regulation of personal information.

In the current climate, individuals and organizations are reluctant to share personal and professional data due to data privacy and misuse issues and increased data regulation, AI organizations with limited resources do not have access to larger valid datasets to train better models, and published models can quickly become out of date without effort to acquire more data and re-train them. As a result, the focus of AI has shifted from an orientation centered on AI-based algorithms to an orientation centered on big data architectures that guarantee security and privacy. Isolation of data and protection of data privacy is becoming the next challenge in AI.

Expensive training costs

While advances in hardware and software have been driving down AI training costs by 37% per year, the size of AI models is growing much faster, 10x per year. As a result, total AI training costs continue to climb. ARK[8] believe that state-of-the-art AI training model costs are likely to increase 100-fold, from roughly $1 million today to more than $100 million by 2025.

Centralization and De-Democratisation

The democratization of AI means that more people are able to conduct AI research and/or build AI-powered products and services, and democratization is the lowering of barriers to entry in terms of resources and knowledge. This includes.

The right to use powerful AI models.
The right to use algorithms.
The right to use the computational resources required by the algorithms and models.
The right to use algorithms and models without advanced mathematical and computational science skills.

While AI has made tremendous progress, the benefits of AI are not widely used, AI has not yet been democratized, and there is a trend toward increasing centralization.

Most AI research is controlled by a handful of tech giants. Independent developers of AI have no readily available way to monetize their creations. Usually, their most lucrative option is to sell their technology to one of the tech giants, leading to control of the technology becoming even more concentrated.
A few tech giants have monopolized the upstream of data by providing services to consumers, gaining unprecedented access to data, training high-end AI models and incorporating them into their ecosystem, further increasing the dependence of users and other companies on the five giants. Except for a few tech giants, other market players such as small and innovative companies find it difficult to collect large-scale data, and even if they obtain data at a significant cost, they lack effective usage scenarios and are unable to exchange them, making it difficult to precisely align with relevant AI learning networks.
Most organizations face an AI skills gap, and the core resource for filling that gap is AI talent, including AI researchers, software developers and data scientists. But tech giants are strategically working to monopolize AI talent at an unprecedented rate and scale, and these companies’ AI developers are available only to advance their employers’ goals.

AI needs Blockchain & Privacy-preserving Computation

Blockchain, privacy-preserving computation, and AI affect and act on data in different ways, and the combination of these technologies can take data utilization to new levels while enhancing the blockchain infrastructure and enhancing the potential of AI.

Blockchain consensus algorithms can help subjects in decentralized AI systems collaborate to accomplish tasks. For example, in the field of intelligent transportation, AI is the “brain” behind countless autonomous vehicles, and these autonomous vehicles need to cooperate with each other trustfully to accomplish a common goal. Artificial intelligence systems have no mechanism to ensure that these autonomous vehicles can reach consensus among themselves in a trustworthy manner. Of course, the collaboration of these autonomous vehicles could rely on trusted third parties, which would expose the public to security and privacy issues.
Artificial intelligence models require massive amounts of high-quality data for training and optimization, and data privacy and regulation prevent effective data sharing. Blockchain and privacy-preserving computation enable the privacy and security controls needed for compliance and facilitate data sharing and value exchange.
The intersection between artificial intelligence and cryptography economics is another interesting area where blockchain combined with AI can enable the monetization of data and incentivize the addition of a wider range of data, algorithms and computing power to create more efficient artificial intelligence models.
Blockchain can make AI more coherent and easy to understand, with an untamperable record of all data, variables and processes used in AI training decisions that can be tracked and audited.