Meeting on the development of artificial intelligence technologies 2019-05-30 15:30:00 Vladimir Putin held a meeting on the development of artificial intelligence technologies. It took place during a visit to School 21, established by Sberbank as an educational organization to train IT specialists. Before the meeting, the head of state was told about the academic process at School 21 and had a brief conversation with students. The President was informed about the school by Head of Sberbank German Gref and school Principal Svetlana Infimovskaya. The students of the school can study the following areas: Algorithms, Graphics, Mobile Development, Computer Security, Robot Technology, and Artificial Intelligence to name a few. The school has 940 students today. Every year 1,000 software specialists will graduate from the school. On average, students are expected to study for 2–3.5 years. The course includes two practical training sessions in relevant companies for six months or more. * * * Excerpts from transcript of meeting on the development of artificial intelligence technologies President of Russia Vladimir Putin: Good afternoon, colleagues. Today I suggest that we discuss concrete steps that will form the foundation for our National Strategy on the development of artificial intelligence technologies. We have repeatedly spoken about the need for such a comprehensive document. I also mentioned it in this year’s Address to the Federal Assembly. This is indeed one of the key areas of technological development that determines and will continue to determine the future of the entire world. The artificial intelligence mechanisms will allow for quick real-time decision-making based on analysing vast amounts of information known as big data, which provides tremendous advantages in terms of quality and performance. In addition, such mechanisms are unparalleled in history in terms of their impact on the economy and productivity, the effectiveness of management, education, healthcare and daily life. However, vying for technological leadership, primarily, in the sphere of artificial intelligence – and you are all very well aware of this, colleagues – has already lead to global competition. New products and solutions are being created at an exponential growth rate. I have said it before and I will say it now: he who can establish a monopoly in artificial intelligence – we are aware of the consequences – will rule the world. It is no accident that many developed countries of the world have already adopted action plans to develop such technologies. Of course, we must ensure technological sovereignty in the realm of artificial intelligence. This is the most important prerequisite for the viability of our businesses and the economy, the quality of life for Russian citizens, security and, finally, our defence capability. Here, we are not just talking about algorithms for addressing individual and highly specialised problems; what we need are universal solutions, the use of which gives the optimum effect in any industry. In order to achieve such an ambitious goal in AI technology, we are objectively positioned to have a good start and we have a serious competitive edge. Today, Russia boasts one of the world’s highest penetration rates for mobile communications and internet access, as well as for the development of electronic services. The cost of internet service in Russia is one of the lowest internationally. We also rely on traditionally strong scientific and education schools in mathematics and physics and a competitive system of training IT specialists. Incidentally, Russian students have won the International Programming Contest for the eighth year in a row. This year, it was a team from Moscow State University. Strong scientific and applied competencies have already made it possible to create original and, most importantly, commercially viable world-class devices in Russia, including in such areas as computer vision and voice recognition. We just saw this when our colleagues showed them to us, as well as cyber security. We have five years to achieve leading positions in these areas. Colleagues, what are the tasks that we need to focus on and prioritise in the Strategy based on Russian and international experience, as well as global trends? I will only list those that I believe should be a priority. I hope to hear from you as well. This is why we have gathered here today. We need to make the Strategy complete and effective. The first priority is to lay conceptually new fundamental groundwork, create mathematical methods and operation principles for artificial intelligence, including those based on the human brain. Russia must become one of the key platforms for solving the most complicated scientific problems involving scientists from all over the world. This work could also be performed by the international mathematics centres that will be opened in Moscow, St Petersburg and Sochi next year, as you may recall. In fact, they will complement the facilities we already have. It is necessary to multiply the funding for artificial intelligence research and create incentives for private investment and corporate science, research and development. We need to discuss this issue specifically. Second. We need to build our workforce and intellectual potential, retain our talent and attract the best professionals from all over the world. There are actually many people working in this industry in Russia, but still much fewer than in the countries that are pursuing leadership in this area. To secure this leadership, we need to develop study programmes in artificial intelligence at universities and colleges, including those in the regions. It is also important that we propose efficient mechanisms for monetary incentives, comfortable working conditions, including remote work, for our programmers, engineers and scientists and for the leading experts working abroad who, as you know, include many Russians, and they are happy to return to Russia and work with their colleagues here. Of course, when necessary, it is important to expedite any formalities related to citizenship, work permits and other requirements. Third. It is absolutely important to bring our legislation up to date with the new technological realities, to move quickly to take quality action to build a flexible and adequate legal framework for developing and applying AI-based solutions, as well as special regulations for private investment in breakthrough solutions and, certainly, to guarantee solid protection of intellectual property and legal conditions for patent registration in Russia’s national jurisdiction. I want to repeat that it is important to lift legislative and administrative barriers for technological pioneering while ensuring the security of the state and society with full observance of our citizens’ rights. There is room for improvement. Fourth. We need to build an efficient legal regulation for data circulation, advanced infrastructure for data processing and storage as well as propose balanced solutions that will allow using data for developing AI algorithms. And, finally, the most important issue is whether our society and people are ready for the pervasive implementation of this technology. It is necessary to provide for mass digital literacy and re-training programmes – especially in sought-after jobs. Today Mr German Gref and I reviewed his school. It is indeed a great example of specialist training and promotion of interesting, promising startups of great significance for almost all sectors. So let us begin with this point. Sberbank CEO German Gref: I am speaking today on behalf of a large group of our colleagues, because what was happening over the past several months since you issued your instruction was a major joint effort, sort of a public-private partnership. What we are presenting today is a major result of the joint work done by a number of our colleagues from the Government and business community. Indeed, it was a major undertaking, because over this period we have drafted not only the document titled “The Strategy,” but, importantly, we also managed to create a road map at the same time. So, today, we have two draft documents which can be considered a complete document that only awaits approval. Today, the word ”technology“ is changing everything around us. If you ask about the most important word affecting politics, companies and the life of every person, the answer will be simple: this is, of course, the word “technology.” Speaking of artificial intelligence, they say that AI today is a gateway to reaching new levels for all technologies without exception. There is not a single technology or a sphere of research that has not taken on a fundamentally new dimension today in connection with the use of artificial intelligence. A mass race for national leadership in the sphere of artificial intelligence began in 2017 when five countries adopted national AI strategies and, in 2018–2019, 30 countries already adopted national strategies for promoting artificial intelligence. In this sense, if the document is approved in the near future, we will be among the 30 countries that have drafted clear road maps and stated that AI is their top priority. Different countries have different goals. The strategy is an interesting document. We have reviewed almost all strategies, including those of the United States of America, China, France, the United Kingdom and the United Arab Emirates. These are fine documents which outline a variety of goals. Universal goals are set in the strategies of two countries – the United States of America and China – which are to retain leadership in the sphere of artificial intelligence, and to become the leader by 2030. Other countries have set themselves the task of becoming leaders, primarily, in developing technology, and become leaders in one area of AI. Nevertheless, we see that this has become the dominant trend today. According to EU estimates, industries that introduce artificial intelligence will grow at an average rate of 9–12 percent per year which is three to four times faster than the rest of GDP. Indeed, artificial intelligence provides ground-breaking opportunities in resolving almost any problem. When it comes to artificial intelligence, the first thing we did, and we spent a lot of time on it, was to determine what we meant under the notion of “artificial intelligence.” There is some debate as to whether it’s artificial intelligence, machine intelligence or a system of machine training. But we decided to leave the definition “artificial intelligence” and included five elements in it. To begin with, this is computer vision, natural language processing, speech recognition and synthesis, recommender and decision-making systems, and, as a separate area, the advanced methodology and technology of artificial intelligence, primarily ML-technology, that is, the technology of automated machine learning. In developing the strategy, we tried to analyse each of these areas of artificial intelligence and determine how to use it in industry, agriculture, transport, public services and justice. Incidentally, our judicial system, the commercial courts, common law courts and magistrate courts have very good experience in this respect. They started introducing artificial intelligence last year. I think users and the public will begin to see this soon. Procedures are expedited while decision-making is becoming more precise and transparent. This applies to almost every department and every area that we analysed. Naturally, with the adoption of this strategy we can ensure, to a large extent, the integrity of using this technology because for now only some bits of it are being used in some areas. If we apply the technology as a whole, its value will increase substantially. And of course, we will be able to seriously influence the quality of services for end users, or everyone in the country. Our goal is to make Russia a leader in global artificial intelligence. Three major goals were specified: increase the wellbeing of our people, create incentives for economic growth, and ensure national security and law enforcement. We discussed for a long time how to reach these goals, what KPI would be used for each of these areas, and we agreed that we are exactly the right country that can pursue these fairly ambitious goals. There are only five countries in the world that have the major elements of a digital ecosystem, like their own search engine, social media, and national email services. We have these key elements. They have been created over the past decade. We are now at a point where individual companies cannot achieve success, but if we pool our efforts – business, the scientific community and the Government, including the Russian regions, we will be able to achieve dramatic changes. We identified and analysed six drivers behind the development of artificial intelligence as we were developing the strategy. We identified and outlined them very clearly. These are the algorithms and mathematical methods, software, data, data management, regulation and use of data, and everything connected with education and personnel, as well as regulations. Each element is critical to this system. The absence of even one creates critical risks for the entire system. Actually, I will quickly go over these six measures and sum up the contents of this document. For each of the areas, we have set goals, algorithms and mathematical methods. We wrote that we want to be in the top 10 countries in terms of the number of papers and participation in conferences by 2024, and in 2030 we want to join the top 10 countries in terms of the average citation rate. An entire set of support measures are outlined there. Research centres and laboratories should be created, special long-term financial support should be set up for researchers, and special computer resources should be created, because artificial intelligence requires its own supercomputer capacities. This is special-purpose equipment which is quite expensive. Here, we will probably need to pool our efforts, because every company definitely does not need to have such a supercomputer centre of its own. We should build, maybe, one or two centres, which almost all the participants will be able to use, and using it 24/7 can make this possible. The second area includes developing software and technology solutions. We have also set quite ambitious and clear goals here. We said we want to develop solutions for special tasks which would be superior to humans. By 2030, we must ensure superiority in a wide range of tasks. The key goal of all AI developers today is to build “general artificial intelligence,” or super-universal, or super-powerful computer intelligence capable of handling any assignment. We have not yet identified these goals in the strategy; perhaps, this will be part of the next stage and the next strategy. We will get there one way or another. Nevertheless, we will have to create the entire technological base ourselves. Here, too, we have outlined the support measures and concrete steps to achieve the second goal. The third goal is storing, collecting and processing data. Here, too, we have outlined measures of support and our goals. We want to create an online platform with anonymised government and company data to which developers of artificial intelligence systems will have access. The next area is the development of specialised hardware. This is probably the most difficult element of our strategy. We can see that we still have not laid the necessary groundwork here, because it is not just hardware, specialised hardware, but also specialised chipsets. Two problems need to be solved: this is the creation of our own architectural capacities that will be able to create the architecture of the respective chipsets, and, accordingly, a specialised production facility that will be able to produce them. This is an area that we will need to address together with industrialists, in particular, with the Russian Technologies company and with other private companies. Here we also determined what needs to be done. Hopefully, we will cope with this task, perhaps, thanks to international cooperation. Another area is personnel training. We said that we want to enter the top 10 countries in educational programmes for artificial intelligence by 2024. And we want to eliminate the shortage of specialists in artificial intelligence by 2030. You cited the figure, Mr President: today we have about 6,000 to 6,500 people working in this area in the country. I can say that Microsoft’s artificial intelligence labs alone today have over 6,500 employees, while 10,000 people are involved in developing Amazon Alexa. Therefore, we have a lot to do here, together with the state, and there is probably also a lot of work for companies to take part in. The last point in our strategy, probably one of the most important, is developing the correct regulations in the field of artificial intelligence. We already have experience from the world, we have the experience of the so-called GDPR’s adoption in Europe, in the European Union, which actually stopped the development of artificial intelligence and there are signs of overregulation in this sphere. Today, all companies are trying to relocate their research centres from Europe, because fines are tied to the turnover of companies; no one wants to be fined up to 10 percent of their company’s turnover, no one wants to test the power of this regulation, so it is easier to create centres somewhere outside the European Union. Therefore, it is important to steer clear of two extremes: not leaving this sphere unregulated, while still creating opportunities for it to maintain the dynamics of its development. And, in effect, there are two key issues being submitted for your consideration today. We ask you to approve the national strategy for the development of artificial intelligence, create a relevant coordinating body and instruct the Government of the Russian Federation to endorse a roadmap for the development of artificial intelligence. Unfortunately, the development of artificial intelligence requires huge investment. If we look at some national strategies we see that each country decided to invest no less than $1 billion in this area. Some countries are investing from $5 to $10 billion from the state budget into this technology. Experimental sites are very important in this respect, as well as communication between research, universities, business and, in our case, the state and the Russian regions. Mr President, we would like to ask you about one more thing. We need experimental sites. Thank you very much for your instruction on the digital education experimental site; work on this is underway. The Ministry of Education is very active in this area. I think this year we will manage to adopt relevant regulations. We have made serious progress in this area just since January. We would like to ask you to consider a special status for a constituent entity, for Moscow, so we can experiment in Moscow on using artificial intelligence systems and testing this technology. In fact, this is what is already happening now because Moscow is very active in this area. A big team is working in Moscow and they are investing a lot in this. Mr Sobyanin (Moscow Mayor) is ready to transfer all the developments that have been achieved in Moscow to the regions at no cost. We think this would be very good experience if we test the systems using Moscow’s data, as this data is collected daily. Huge amounts of data have been created and the data processing systems have been developed. This would help all the regions in the country make serious progress and would prevent us from wasting a lot of money throughout the country. Experimental sites, then scaling and probably, the localisation of this product in several regions, and later, nation-wide expansion. These are the decisions we would like to ask you to consider, Mr President. Deputy Prime Minister Maxim Akimov: After this bright picture and a fairly detailed description of our truly ambitious plans I will probably add some organisational brush strokes to this picture. I would like to say straight away what we are not going to do at all, namely to engage in yet another bureaucratic construction. We believe the organizational tools have been created under the Digital Economy programme. By concentrating the resources that are now allocated for this programme within the framework of a separate federal Artificial Intelligence project, we will be able to work with the strategic challenges described by Mr Gref. We are not starting from scratch. Even prior to this phase, we were actively working on the road map under the cross cutting technology project with Sberbank’s leading and coordinating role. What measures could this plan include with total funding of up to 90 billion rubles for the next six years in our estimate? Naturally, first and foremost, this is support for research in algorithms and mathematical methods, including support for the leading research centres. It is necessary to create a favourable environment for developing and expanding technologies, and to subsidise pilot introductions, because they involve a risk that private companies may and should share with their public counterparts. Therefore, we will earmark resources for leading companies to create prototypes for using artificial intelligence, thereby creating a foundation for future applications. As mentioned earlier, it is not enough to create a proving ground for making high-quality artificial intelligence. We mainly need to collect a package of information that should be as diverse as possible, so-called libraries on all potential situations. This cannot be done in closed conditions. We could indeed rely on the leading regions using not only financial and technological opportunities but also opportunities in the area of data turnover and concentration. That means, primarily, Moscow that can work with experimental regulation documents. As mentioned earlier, data is food for artificial intelligence. Therefore, one of the tasks is to cover as much impersonal information as possible, while observing confidentiality requirements. In cooperation with the Federal Assembly, we have drafted amendments to the law on personal data, which specify mechanisms for depersonalization. This is a large area of work. We hope this law will soon be considered by the Federal Assembly chambers. It is also necessary to update industry standards, including those in security. I am now moving on to practical application. Here is a very relevant example: predictive analysis of complex equipment, pressure equipment, commercial sensing, commercial logistics, commercial power generation, which, by the way, will create opportunities for transitioning to a risk-oriented approach to control and supervision. Of course, we are not just talking about personnel training but about changing the managerial and partially engineering culture. In the long term, all jobs that involve serious decision-making will require artificial intelligence competence. We will need to cooperate with the Ministry of Science and Higher Education and the Ministry of Education to substantially review education standards. We propose appointing the Ministry of Digital Development, Communications and Mass Media the federal executive body in charge. Working with Sberbank in its leading role and with an extremely representative group of technological companies and competence owners is a good example of how work could be organised in other areas of technological development. As per your instructions, Mr President, we are moving fast in this direction because Russia’s and global experience do indicate that the largest companies play the most important role. For example, in the past five years, up to 80 percent of patent applications in the area of artificial intelligence were filed by corporations rather than research or academic centres. It is Russian big business that must and can become a major actor in developing high-tech industries in Russia. As you may recall, as of now, it has been proposed to identify 13 high-tech industries. They include artificial intelligence, a new generation of mobile wireless communications (primarily 5G Plus), a new generation of narrowband wireless communications for the Internet of Things, short-range and medium-range communications and blockchain technologies. As agreed with the initiators, we divided the quantum industries into three tracks, including quantum sensors, quantum computing and quantum communications. There are also new material and substance development technologies, satellite communications, a new generation of microelectronics and development of an electronics component base, development of new portable power sources and their transportation as well as genome technology. I apologise for the long list but it was important to give you the overall picture. We are suggesting a format where the first step would be the signing of a letter of intent between the Government of the Russian Federation and the involved companies. This will set not only the general outline but also commitments on developing and adopting detailed roadmaps for each high-tech area through 2024. The text of this letter of intent is ready and we have discussed it with the companies that are here: Rosatom [State Atomic Energy Corporation], Russian Railways, Rostec [State Corporation for Assistance to Development, Production and Export of Advanced Technology Industrial Product], Rostelecom and Sberbank. We will be ready to sign it soon, in June. By agreement with the companies, the following distribution of themes is suggested. Sberbank will be the lead company on artificial intelligence; the Rostelecom and Rostec consortium will be in charge of relevant parts of the 5G mobile communications technology; Rostecnologii will supervise the development of quantum sensors, blockchain technologies and a new generation of broadband technology for the Internet of Things; Rosatom will be in charge of quantum computations and new materials; and Russian Railways will deal with quantum communications. I would like to make the point that a lead position does not imply a monopoly in a particular area. This is borne out by the practice of work on the first area – artificial intelligence. The Government of Russia will monitor the involvement of the companies in the process. I would like to note that the Russian Direct Investment Fund (RDIF) is already making a tangible investment in startups and more mature companies involved in artificial intelligence technology. A strong team can only be established by forming a large consortium. <…> Russian Direct Investment Fund CEO Kirill Dmitriev: Investment is vital for developing artificial intelligence and is a priority for the RDIF. I would like to briefly describe our proposals in this area. First, we see that artificial intelligence can raise the effectiveness of many industries by 30–40 percent. There are examples of reducing energy consumption by 40 percent in data centres, increasing the detection of lung cancer by one third, and raising the harvest yield in India by 30 percent. This 30–40 percent increase in efficiency is extremely vital and a key driver. Many consultants believe that the introduction of artificial intelligence will lead to an annual GDP growth of 1.2 percent and produce a total of $13 trillion. Russia has very good starting positions. We have analysed 100 leading AI companies, selected the most promising 20 and approved investment in six of them, including Oncobox, which is an innovative system for choosing proper cancer medication, and Motorika, which develops prosthetic devices. We are also introducing artificial intelligence in our portfolio companies. Introducing intelligent systems in our cancer treatment centres will reduce the number of medical errors five to six times. In conjunction with the United Arab Emirates, we are launching ScanEx in Middle Eastern markets. Our partners have allocated $2 billion as a joint investment in our projects. Of course, we can do some of the projects without our foreign partners, but where it makes sense to build a consortium, we have access to the capital. Here are eight short points that I would like to make. First, we see that there are two big data management models. One enjoys more recognition in China, where the state provides greater access and controls data, while the other is more European in nature, where access to data is very limited. We, of course, believe that the Chinese model will allow us to move forward faster. It is also important to raise Russian companies to the level of international leaders. If companies work only for the Russian market, they do not have enough of a competitive edge in order to compete internationally. We want to bring our companies to international markets. The third initiative is that it is, of course, important for our companies to implement AI projects. We believe that each major state-owned company must have its own strategy for implementing AI projects. The fourth is that it is possible to build consortia with China and the Middle East whenever it makes sense in order to have access to large-scale markets. Fifth. Mr President, you noted the appealing environment for specialists, and we plan to create an artificial intelligence centre in conjunction with Moscow State University. The sixth is data centres. Seventh. We already have a lot of data in Moscow, including the Federal Taxation Service, which offer significant potential for analysing the economy. These data can be put to good use. Gazpromneft CEO Alexander Dyukov: I have several points to make regarding the Artificial Intelligence Development Strategy. Firstly, I think we have drafted a solid basic document, which takes into account the key development areas. We contributed our suggestions and comments during the strategy development process, and most of them were taken into account. At Gazpromneft, we are already working in many of the areas identified in the Strategy. Secondly, of course, key factors such as technology, algorithms, data, people, and infrastructure are needed if we want to develop artificial intelligence and use it in real life. However, the demand for artificial intelligence technologies is another fundamental factor underlying development. The demand needs to be shaped and consolidated in order for the AI development strategy to make sense. Informed demand is currently available in areas such as banking, media, retail and telecom, but the volume is limited. Regarding the manufacturing industry, we can say that our industry, with the exception, perhaps, of Rostec and a number of other companies, is, in fact, not seriously involved in creating demand for AI technologies. This is not criticism, and there are indeed objective reasons for this. This situation pertains not only to this country, but to the rest of the world as well. However, the manufacturing industry accounts for a significant segment of the economy, and it can and should be one of the drivers behind economic growth. Industry has its own specifics in the sphere of AI use, and we cannot rely on the fact that developing artificial intelligence, say, in telecom, will allow us to resolve the issues facing the manufacturing industry which has its own peculiarities, and this must be taken into account when implementing the strategy. This also requires special attention on behalf of the state in terms of the availability of the requisite legal framework and infrastructure. I believe that involving the manufacturing industry is important for developing artificial intelligence. The fuel and energy complex – the largest segment in Russian industry – can play a critical role here. The fuel and energy complex has a number of key advantages over other industries that are necessary to becoming a leader in developing artificial intelligence. The fuel and energy complex can form solvency and demand for technologies. The introduction of artificial intelligence technology in the fuel and energy industry will have a large-scale economic effect. The fuel and energy complex boasts strong managerial and engineering personnel. Also, the fuel and energy sector is one of the most complex ones in terms of the volume of data created during production activities and the variety of tasks and business processes. Therefore, the tasks addressed by the fuel and energy companies can then be scaled and replicated in other industries. Regarding Gazpromneft, we have already begun developing artificial intelligence and using it to overcome the challenges facing the oil industry. Our company has gained experience in using artificial intelligence in many areas from geological exploration to petrol stations. Since my time is limited here, I will not dwell on individual projects. I just want to point out that our digital transformation programme already includes 12 key programmes and over 500 projects along the entire added value chain. We need to create hundreds of models within these programmes. According to our estimates, the economic effect will result in at least an additional 150 billion rubles of operating revenue annually. Mr President, we are ready to provide our expertise and experience, as well as the production base for developing artificial intelligence in the manufacturing industry. We are prepared to work in conjunction with other companies and organisations on implementing the Strategy. We are also willing to assume a leadership role in developing artificial intelligence for the manufacturing industry. As I said earlier, this is very important. Rostec CEO Sergey Chemezov: Regarding the first question, Mr Gref listed six factors of AI development. One of them includes developing special hardware. We have already started working on this and have created a public-private partnership with AFK Sistema. We have combined the assets that are in one way or another connected with this area of business. Our main goal is to create a modern electronic component base, primarily chips, that can be used later to create AI hardware. Regarding artificial intelligence in general, yes, I want to emphasise that this enterprise remains open, and we are ready to take aboard anyone who is willing to develop this area of activity and who has something to offer. I know about Angstrem T. In the future, I think this company will be able join our community as well. As to the overall view of the other fields in the AI area, all our companies, many companies have already been working in this field. This includes face recognition, one of our companies also has been involved in that; drones, airplanes, helicopters, driverless vehicles – those designed at KamAZ are already operating at least around the KamAZ grounds. I hope the day is close when this vehicle will hit the roads. There is an automated system for radio electronic warfare with drones, which not only stops a drone approaching a protected facility, but also lands it or turns it around and sends it back. Within a radius of three kilometres, it covers any object with its dome – overall, it is a pretty amazing thing which is being used among others by the Defence Ministry. I am not even mentioning the things we already have in our defence system – self-homing missiles, they have been in operation for a long time and are actually artificial intelligence as well. Mr Akimov has already named a number of hi-tech fields, designing a fifth-generation wireless system being one of them. Deploying 5G networks is a key challenge in developing Russian telecommunications, and the solution requires the combined efforts of mobile operators and industry both to clean up existing frequencies and to design and subsequently introduce critical components of 5G networks. Rostec is actively engaged in consolidating the operators and vendors for the development of 5G wireless communications in the form of a consortium. The corporation is drafting a roadmap for wireless communication technology development, and alongside with it we established, on Mr Akimov’s instructions, a concept for designing industrial solutions for 5G networks. As of today, the approaches outlined in the concept have been approved, the concept has been coordinated with Russian industries and is being actively discussed with mobile communications operators. I would like to stress that the introduction of 5G technology would require domestic solutions that will ensure high quality and uninterrupted operation of all services, free from the impact of external factors. This is crucial. As such, it is logical that a technological infrastructure should be created first, with all services to be built around it later. This is why I would like to offer Rostec Corporation’s services so that we could head up this field together with Rostelecom. Deputy Prime Minister Yury Borisov: Regarding the Strategy. After considering the document I should note that it is timely, relevant and fairly well worked out. Therefore, I express gratitude to all the authors since practically all aspects have been addressed, all the six tasks presented to us are urgent and definitely require a comprehensive solution for us to achieve the final goals. A few words about the target. I would like to see in the Strategy, alongside with undoubtedly important goals such as joining the top 10 countries in terms of the number of papers and citation rate, also more tangible goals regarding the market share, both global and domestic. I think the Strategy’s implementation priority is the rapid introduction of domestic – again, domestic – artificial intelligence solutions. And this concerns precisely that set of algorithm, programming and hardware solutions in all areas of the economy with a view to dominating the domestic market of these products (and estimates show that it is a colossal market) and also establishing positions on the foreign market. I think that the implementation of this strategy, especially concerning hardware support, will create a synergy effect for the development of national microelectronics, because I have first-hand knowledge of these products. They are complex, with high added value. Unfortunately, the sales period is long and they are fairly expensive to make. Without a doubt, the measures provided for in the Strategy will be directed at creating such products. I think the key task is promoting the products on the market. I fully and completely agree with Mr Dyukov that generating demand is the key priority of the Strategy. Taking into account that these products are expensive, it will probably require considering a state order to introduce these products at the specialised centres Mr Gref spoke about. Their number should be limited. But domestically manufactured products are competitive and must break through on the country’s market. These are the two remarks I would like to be addressed. Vladimir Putin: I am totally with you on that. Because we must strive for the results of our work to be quantifiable, though the targets we set before us may include those that at first glance are not really tangible. This is true. Yandex CEO Arkady Volozh: Our specialists were also involved in the development of the programme. Much of what we advised was taken into consideration. I would like to point out two aspects that require more attention. The first is science, education and personnel. The second is testing, introducing AI-based solutions and regulatory parameters. It has been said here that more than 30 countries have their own artificial intelligence programmes. The reason we have things to discuss here today is that Russia has its own specialists in this field, and Russia is actually one of the unique markets in the world in terms of introducing these systems into the market, the retail market – we have almost everything domestically. This is possible because we have our people, and our own system of selecting and educating these people. It would be good if we could carefully preserve what we have, and not just nail it down at the university level, which is being done now, but also – this is stipulated in the Strategy – we need to go deeper and begin a systematic selection and training of specialists in exact sciences, mathematics, and programming right from school. This is one aspect. And the second aspect is, when we train people, it is no news that Russia is a donor country in terms of such specialists for global companies; it is normal for young people to go to global companies after training. It is important that some have an incentive and a desire to stay, or to come back to work for Russian companies. At Yandex, we have brought about a hundred specialists back home over the past couple of years, they have returned from global companies to work with us. We attracted them, not so much with high salaries actually as with the opportunity to solve ambitious tasks and to develop projects that are being applied right before our eyes, and people around you are starting to use them. This, as a rule, is not the case when they work at other global corporations. It would be desirable to have a special state programme within the Strategy that can provide an incentive for people to return to work here – not so much a financial incentive, perhaps, but something intangible maybe. Other countries have such programmes, and we also need to have one. The second aspect is creating conditions for testing systems. Artificial intelligence implies machine learning, and machines need to be trained with real data, in real life situations. For example, drones – unmanned vehicles – that are designed to navigate roads should not spend all their time in training grounds, but be tested in the real world, and it is very important that this area should not be overregulated. It is clear that everyone is a little wary, but I would say that, if we do want it to be implemented faster – and this is a very competitive environment – we need to decrease it and relax a little bit here, and enable very quick implementation, and a chance to test a large number of machines or any AI-based solutions in real conditions. Yandex needs to put a hundred vehicles on the streets this year. If we use the current procedure, certifying these vehicles alone would take four years. It would be good to reflect this in this programme – testing in a real environment. This is about whether we will import or export these technologies in the long run. Special Presidential Representative on Digital and Technological Development Dmitry Peskov: I would like to follow up on what Mr Volozh said and discuss the issues he raised. First, I can note with satisfaction that much preparatory work has been done as part of the national technology initiative and its results were later included in the strategy and used by our colleagues. Second, the area of development, which we now have, but which, probably, is reflected insufficiently, is connected with the data types that we work with. Today, we rely on the data which has already been created by banks, telecom companies and the state. However, there are much larger data sets as well, and we see that the world is making a turn towards completely unconventional data types such as oceans, forests, humans, biome and microbiome. We see vast numbers of start-up companies that are competing for traditional markets but based on a relationship between biology and artificial intelligence. Innovative products are showing up on the market. As you may remember, we had a major discussion about the product labelling system. Now, we have a company that is building a global labelling system based on a unique footprint, that is a biological imprint of the place of production. I believe this relationship should be finalised and a DataSet formed not only using banks and telecom companies, but it should also be used in articulating tasks in scientific research, various industries, such as the forest industry, and many other areas from the point of view of forming new DataSet types. Otherwise, we will succeed in doing our homework to narrow the gap, but lag behind during the next step. Second, issues related to personnel and legal regulation. The gap in personnel needs is currently so catastrophic that we were unable to find a single scenario where we could catch up even by 2030. In this sense, if we continue to focus on the idea that we can modernise the existing system of educational standards, we will never get the result we are looking for. The relationship between technologies is changing so quickly that as soon as we enter this very long bureaucratic cycle, we will have trained new types of specialists en masse by about 2030. We went over this topic many times, but things are already overdue now. We need a separate regulatory territory on the subjects of artificial intelligence, big data and other end-to-end technologies which would allow us to launch the training process within two to three months, rather than three, four, five or six years. Again, there is not a single scenario available. The problem is that the data is really what provides the feed. The data gives rise to many new specialties. Today, they are not reflected in any single regulatory document. Fortunately, the good news is that they create many new jobs. In this sense, we are not fretting that artificial intelligence will take all the jobs and no new jobs will be created. However, the hard part is that deeper division of labour calls for separate regulations. Please have them outlined by issuing additional instructions, as they have not been formed in full. My final point is that, of course, the entire education system has to be changed. Today, we are taking such a step in conjunction with our colleagues from the Ministry of Science and Higher Education. We are beginning to implement an AI-based education platform at a hundred regional universities and we hope that it will start in July. I invite everyone to attend the launch. <…> First Deputy Director General of Rosatom Ivan Kamenskikh: I would like to say that Rosatom has historically dealt with this subject from the outset. In fact, this is about automatic control systems (ACS) for technological processes at nuclear power plants and core defence activities. We have very close ties with Rosneft and Gazprom when it comes to digital wells and digital reservoir models. We are active in this area as well. We agreed to work with Rostec in supercomputers and mathematical simulation. In other words, this area is, of course, supported. I would like to ask you to support what Mr Gref said today. I would also like to thank Mr Akimov and Mr Belousov for their work in organising things. Without their involvement, we probably would not be able to meet here on such short notice. I would like to support Mr Borisov in that our most important goal is to create an internal market and external market for implementing these technologies. Presidential Aide Andrei Belousov: By way of comment, what Mr Kamenskikh and Mr Borisov said about the market… Colleagues, of course, we focus on all of this, and the entire national technology initiative focuses on the markets. However, the market for AI products is, first, non-existent and, second, is by far not the most important thing. What really matters are changes in the traditional markets which come with artificial intelligence. We should not measure progress by our share of the AI market, but by how much of the market we will be able to win over in the manufacturing industry, and the trade and logistics services. This should be our measuring stick. Operationally, of course, these are not quantifiers, and we will never be able to measure this. We mean this, knowing that this is the main effect, but providing any numbers for 2024 or 2030, you understand, will be on the verge of speculation. I want to repeat again: Mr Gref showed us a slide where we specifically identified the areas where ordinary products cannot be promoted without artificial intelligence. For example, what is a fifth generation aircraft? It is a new concept, a new combat concept, when data of different qualities is combined in real time and artificial intelligence helps you to make decisions. This design will not work without artificial intelligence. The same applies to logistics, commerce and healthcare. Same designs. These are the markets we are talking about. And this is where real breakthroughs are possible. Second. Speaking about problems with sales, and so on. They will continue to occur. Today we must create excess supply, and we must do it deliberately. What is the reason for this sales problem? We must understand that implementing artificial intelligence fundamentally changes the decision-making process in companies. It is impossible to implement artificial intelligence without changing the management system. It will simply not work. The principles and approaches are completely different. Not just in this country but in the rest of the world as well. You are well aware that the Americans have set up special research centres to analyse changes in management caused by artificial intelligence and related information and digital technologies. Therefore, here, of course, these processes are sluggish and rather slow, including in corporations. To spur on all these processes – we must, of course, use incentives, too – but most importantly, we must offer a solution that will give the same companies or their competitors, including disruptors, small companies, an opportunity to win back large market segments based on these decisions. It is a slow and lengthy process, as our experience shows, but it is imperative that we go for it. The third thing I want to say is that the biggest mistake we could make is to start building some new management systems, a new framework for managing this process under the Strategy and fall into a rather common pattern for us, unfortunately, with a lot of bureaucracy, various resolutions, decisions and so on. We need to launch everything very fast. This is why it was proposed to rely on the chain of decisions presented by Mr Akimov. It is an existing decision-making system established within the Digital Economy programme. We are making changes to it that will to a great extent affect the contents but will have hardly any effect on the bureaucratic procedure, in order to save time. These are my three comments. Vladimir Putin: Thank you. I want to base my comments on what Andrei [Belousov] said. I agree with almost everything. Of course, as we begin to use artificial intelligence on a large scale, we need to make sure our companies win both national and international markets. At the same time, there are areas such as the application of AI-based technology in medicine, for instance, in organising doctors’ work. To say nothing about the pharmaceuticals industry, where there is plenty of room for incorporating AI. Today we mentioned that efficiency in this field has increased by 30 to 40 percent thanks to artificial intelligence. We can apply it but we do not have to. You see, we do not have to. We were fine without it. But implementing this technology requires certain decisions from ministries and other government agencies. We should promote these products. A similar situation persists in many other industries. There, too, a great deal depends on ministries, including the Ministry of Industry and Trade. We can either establish certain requirements for these innovations or, if we do not, no change will happen. Yes, it could be more expensive if we force industries to do it, to put it bluntly. It may be more expensive at the start but the resulting breakthrough will be so dramatic that the initial investment will pay off. All this requires very careful and thorough consideration. I am sure you will produce a good strategy. I just want to point out the fact that in general, we can draw up strategies, even the most complex ones – and this strategy is complex. We need a step-by-step plan for putting these strategies in action. In this case, we need a step-by-step plan for developing artificial intelligence. This is very important. Mr Akimov has already talked about this but we need to make sure there is a clear idea of how we proceed. Finally, as concerns our major companies that are ready to work with hi-tech technology. That is great. This is not the first time we have discussed this. I want to thank those who participated. I can see we have specific questions and specific companies and, most importantly, there are agreements with these companies for the areas of their interest. But the list is not exhaustive. We have so many more companies and it must be more difficult to get them involved. But please continue working with them and tell me who specifically I need to address and regarding which issues. Maxim Akimov: Of course. Vladimir Putin: Thank you very much.