Artificial Intelligence and the blockchain – The perfect match from Codingcompiler. Artificial intelligence and blockchain are among the most progressive technologies that promise to radically change the life of society. What are the prospects for the integration of artificial intelligence and blockchain? A pair of “blockchain – artificial intelligence” is able to provide fundamental changes in all aspects of society, promising the growth of the global economy, despite the fact that they have a long way to go.
Artificial Intelligence & Blockchain
Artificial intelligence and blockchain today are powerful engines of innovation that are making fundamental changes in all aspects of society and promise the growth of the global economy. The future has already arrived: there are robots and polite artificial assistants who are able to make appointments on your behalf, leading natural and quite meaningful communication. And with the advent of new interaction platforms, participants no longer have to trust “unreliable” intermediaries like Facebook, Yahoo or Equifax. What are the prospects for combining these two technological areas?
Related Article: [ Bitcoin Mining ]
Artificial intelligence is any task performed by a program or machine and requiring the presence of “intellectual” abilities. Artificial intelligence systems have capabilities that are associated with human intelligence, such as planning, learning, solving logical problems, as well as social skills and creativity.
Interest in artificial intelligence has resumed today thanks to breakthroughs in the field of machine learning, especially deep-seated, and also due to the rapid growth in the volume of available data, which contributes to an increase in the efficiency of teaching models.
Related Article: [ What Is Bitcoin ]
There are developments like self-driving cars and robots, and anxiety is also growing. Artificial intelligence also has reverse sides – from the possibility of falsifying news with quite realistic snapshots, sound and video recordings until violations of the right to personal secrecy. There are also issues related to the monopolization of artificial intelligence capacities by a small group of players, such as Google, Microsoft, and Amazon, who prevent others from accessing data, specialists, and computing resources.
Related Article: [Artificial Intelligence Vs Augmented Intelligence ]
Blockchain is an open distributed registry, shared by agreement by all users on the network. Records, such as transactions, are stored in blocks along with hash values and time stamps. Each block is associated with the previous one, as a result of which a chain is formed. One of the key features of the blockchain is immutability: there is practically no opportunity to change any information without the consent of the entire network.
Depending on the consensus protocol (method of creating blocks), the blockchain technology is divided into two groups. In blockchains with proof of work (proof-of-work, POW), such as Bitcoin and Ethereum, users (miners) create new blocks by solving (mining) a computationally complex problem.
Related Article: [Blockchain Technology ]
The miner who received the right to create a block receives a fee and a transaction fee. Work on POW protocols requires a lot of electricity. In addition, miners may be subject to “majority attacks“, during which the reward for the block is reduced; examples are recent incidents with cryptocurrencies Bitcoin-Gold, Verge, ZenCash, etc. The blockchains of the new generation use protocols based on the proof of ownership (POS) that do not require an energy-intensive mining process. In them, the chances of participants to create a block increase along with the number of coins – their share.
The most well-known among this group of blockchains are Nxt, Peercoin, EOS with delegated share evidence, the “clone” of Bitcoin Ouroboros, iChing, and Algorand — developed by MIT based on POS and the algorithm for solving the problem of Byzantine generals.
If we discard the current hype around cryptocurrencies that distract from the real potential of the blockchain, the relevant technologies can actually become the basis of a new serverless Internet and a decentralized World Wide Web, in which users gain control over their data and identity. The blockchains promise a revolution in the health care system – giving everyone control over the use of their records. Moreover, there is the likelihood of more advanced alternatives to virtually any platform known today, including Facebook, eBay, Uber and Airbnb.
Blockchain has many weak points, including in terms of security, scalability and efficiency, and artificial intelligence, in turn, suffers from problems related to the possibility of privacy breaches, as well as a lack of trust and the ability to explain the principle of action.
The combination of these two technologies seems inevitable – they could complement each other to create a fundamentally new generation of digital systems. As shown in the figure, the blockchain will eliminate the need to trust artificial intelligence, ensure privacy and explainability, and artificial intelligence will allow building machine learning systems based on the blockchain, which will be distinguished by enhanced security, scalability and more efficient personalization and control capabilities.
Artificial Intelligence Blockchain
Blockchain can be used as the basis for decentralized trading platforms and coordination platforms for various components of artificial intelligence, including data, algorithms, and computing power. This could contribute to completely new levels of innovation and the scale of the use of artificial intelligence. The blockchain will also make such decisions more transparent, explainable and trustworthy. Given that all the data in the blockchain is publicly available, artificial intelligence will become the basis for providing protection against forgery and privacy.
Related Article:[ Artificial Intelligence Trends Forecast]
Secure Data Provision
One of the driving forces behind the development of artificial intelligence has become the huge amounts of data that are potentially available to researchers, developers, and merchants. The data is the “gold” of the digital economy, but access to them is blocked by high barriers.
First, it may not be easy to get enough data for your learning models if you are not working for someone from online giants like Facebook or Google. In this connection, constructive competition between researchers is significantly limited. Secondly, there are growing problems of privacy – already today there are many cases of leakage and abuse of personal data. Recall at least the recent scandal with Facebook, when data from 50 million social network users were collected without their consent by Cambridge Analytica.
The blockchain provides transparency and data visibility, allowing you to find out exactly which user information was accessed, when and who accessed it. Since the blockchain returns control of data to people, users will be able to share it with more confidence, knowing that the data will be used without any violations for personalization or other useful applications.
Doctors and scientists will be able to get access to huge volumes of anonymized medical records, significantly speeding up the search for drugs and the development of new treatments and medical procedures. There will be more hope in patients with rare diseases, as physicians will be able to study data on similar cases around the world, and in part this idea is already beginning to materialize.
Your Data – Your Price
In addition to control over your data and the ability to share them, blockchain technologies will allow you to sell them using smart contracts, opening up the prospect for creating trading platforms that are more secure and private due to the lack of intermediaries.
Such platforms will reduce the number of barriers for small players to enter the market, leveling the playing field and thus promoting the development of innovation. Using evidence protocols with zero disclosure, businesses and researchers could find the information they need without knowing the details about the data and the identity of their owners.
The ability to detect and filter the necessary data while preserving user privacy is very important. Probably, thanks to this, a new stage in the evolution of artificial intelligence will even begin. For example, the startup Nebula Genomics already offers a trading platform, which connects people who want to go through the sequencing of their genome, and organizations that need this data. And Longenesis has developed a platform that allows for sharing medical data, including case histories, information about symptoms, etc., for money.
Related Article: [ Binary Large Object Storage ]
Selling unused computing power
The blockchain will provide an opportunity to use more distributed computing power for artificial intelligence in a decentralized market – clouds. Developers can use millions of graphics processors owned by gamers to prepare, simulate, train, and deploy machine learning algorithms. Gamers whose video chips are used in games for only a fraction of the time can, with the help of smart contracts, set machine time for sale.
The idea of selling unused computing power is not new; it suffices to recall the once popular grid technology. But the market for cloud services with artificial intelligence functions has a lot of possible applications – all that artificial intelligence is capable of. Encouragement in the form of payment by cryptographic tokens will encourage a growing number of users to sell their computing resources.
Related Article:[ What is Big Data and BigTable ]
Coordination of non-trusted devices
Artificial intelligence has great promise for controlling devices that do not trust each other. Groups of robotic devices, subjects of the Internet of Things and various wearable devices can now coordinate their actions and make collective decisions. In such scenarios, the blockchain serves as a coordination platform, the security of which can be compromised by attackers only if they take control of most devices.
This opens up a large field of applications – from updating the “firmware” of the refrigerator to coordinating the actions of groups of combat robots. True, the same principle can be used to control botnets. Now security specialists are able to disable botnets immediately after the discovery of their control centers, but if this single point of failure is replaced with a blockchain,
Related Article: [ Amazon Machine Learning ]
“Explainable” Artificial Intelligence
Despite the successful use of machine learning in the development of autonomous systems capable of perceiving information, learning and acting independently, the field of its practical use does not expand too quickly. One of the reasons is the difficulty of explaining exactly what processes occur during training.
Decisions made by these systems are not amenable to a strict explanation, and therefore they cannot be verified and trusted by them. With even greater apprehension, artificial intelligence is viewed in medicine and financial planning, where expandability is especially important, due to the fact that the wrong decision can lead to serious consequences. Therefore, it would be important to have an immutable record that allows you to track the movement of data and all the actions of complex systems based on artificial intelligence.
It is this opportunity — to register all the “bends” of the data processing chain on the way to a solution — and gives the blockchain. By tracking the behavior of artificial intelligence systems when processing various input data for different tasks, you can get a more accurate understanding of the decisions made by these systems and trust them more.
At the same time, it will be much easier for people to get the rationale for decisions made by the machine by viewing a detailed log describing the process. In addition, such records will help to optimize the “black boxes” of artificial intelligence, keeping a balance between performance, forecast accuracy, and expandability. And when investigating incidents, the trace recorded in the blockchain will allow to identify the perpetrator.
Related Article: [ Machine Learning, Deep Learning, AI ]
Artificial intelligence for the blockchain
The development and operation of the blockchain require tuning thousands of parameters and trade-offs between security, performance, decentralization, etc. Artificial intelligence can facilitate appropriate decisions by automating and optimizing the blockchain’s performance for higher performance and improved control. And given that all data in the blockchain is publicly available, artificial intelligence will play a key role in ensuring the privacy and privacy of users.
Related Article: [ Blockchain Opportunities ]
Security and Scalability
If the attacker does not own most of the mining resources, it is almost impossible to crack the blockchain, however, applications built on the blockchain platform are unfortunately not so secure. For example, The DAO, one of the largest crowdfunding platforms that owns $ 150 million in Ether cryptocurrency, was the victim of a $ 50 million embezzlement. under the contract fund.
Given the rapid progress of machine learning, the blockchain, running under the control of an intelligent algorithm, could easily detect the fact of an attack and automatically activate defense mechanisms. And if damage is inevitable, artificial intelligence could at least isolate the attacked component from the rest of the blockchain platform. By controlling the blockchain, it could improve scalability and increase the reliability of the distributed registry.
For example, with a sharp increase in the volume of transactions, artificial intelligence could automatically increase the speed at which blocks were created, which would increase the throughput at the cost of increased confirmation time.
Related Article:[ What Is A Fork, Soft Fork And Hard Fork ]
Privacy and Personalization
If you are afraid that regular elections will be compromised, or worried about the security of your data on social networks, blockchain is for you: he will be able to regain control over personal data. But you have to pay for it. In traditional centralized systems like Facebook, Netflix and YouTube, the collected user information is analyzed to personalize the content.
It is thanks to this that when you go to Facebook, you see posts from friends with whom you communicate more often, or when opening Netflix, you see movies that match your taste. The return of privacy means that no one will know your tastes and you may have to scroll through many pages in search of relevant content, without counting on the participation of the automated selection system. But is it possible to provide both privacy and personalization?
Artificial intelligence can come to the rescue – with a new content selection model. A decentralized content provider, such as a blockchain-based social network, could use artificial intelligence to personalize the issue on the user side. The machine learning system runs on the user’s device, analyzing information about the sites it has viewed. Relevant content, in this case, will be loaded by the user device itself, and not “imposed” by the server.
In this case, all calculations will be performed locally – personal data will not leave the user’s device. Moreover, it is possible to “reorganize” the content preferences of the user in order not to allow providers to profile content. Thus, such a model simultaneously provides both privacy and personalization.
Related Article: [How Does Blockchain Work]
Rise of the Machines
The plot with the uprising of machines, taking control of all aspects of human existence, has long been imposed in the teeth, but such a future has already come, at least in part. The blockchains are smart contracts that manage business relationships between users, deciding which of them should get paid. True, for the time being they are not so clever and can only cope with fairly simple conditions.
But soon, artificial intelligence will be able to act as an arbiter in more complex situations. Over time, users will be able to resolve conflicts occurring both inside and outside the distributed registry, without going to court — having obtained the certificates and necessary documents, artificial intelligence could perform automatic arbitration without prejudice and the possibility of influencing this process.
Related Article: [ How is Ethereum Different From Bitcoin]
Despite the rapid progress of artificial intelligence and the blockchain, they still have a long way to go. Of the latest developments in the field of artificial intelligence can be noted Google Duplex – a system that can automatically make phone calls on your behalf. However, for the time being, it provides only three possibilities: booking restaurants, signing up to beauty salons and finding out the opening hours of stores. And, despite the colossal breakthroughs in the field of algorithms and artificial intelligence data, it is still far from the complexity of the human mind. As for the blockchain, recent security incidents with Ethereum, ZCash, Bitcoin Gols, and many other cryptocurrencies suggest that it will take a long time before protected scalable blockchains for real-world applications appear.