Artificial Intelligence: A European Perspective✎
TABLE OF CONTENTS
Part 1: Introduction and AI international landscape
1. Motivation and objective of this report
2. About AI
- 2.1 A brief history of AI
- 2.2 Recent developments in machine learning
- 2.3 Recent developments in social robots
- 2.4 Current challenges
- 2.5 Summary and conclusions
3. EU in the AI competitive global landscape
- 3.1 The international policy context
- -- USA
- -- Other countries
- 3.2 Analysing the key features of the AI landscape
- 3.3 Overview
- 3.4 Technological capacity
- 3.5 Summary and conclusions
4. AI in the EU
- 4.1 Strategies and plans
- --The European Union
- --United Kingdom
- --Other EU countries
- 4.2 Summary and conclusions
5. The AI ecosystem in China
- 5.1 China’s economic and policy context
- 5.2 Government policies and initiatives
- 5.3 Regional/local initiatives
- 5.4 The specificities of the AI innovation ecosystem in China
- 5.5 Summary and conclusions
Part 2: Multi-dimensional perspectives
6. Ethical and societal perspective
- 6.1 Introduction
- 6.2 Overview of individual and collective implications of AI
- --Challenges at individual level
- --Challenges at societal level
- 6.3 Summary and conclusions
7. Legal perspective
- 7.1 Main legal challenges identified in European AI strategies
- 7.2 Ownership, access and sharing of data
- 7.3 The protection of AI inventions/creations by intellectual property rights
- 7.4 Regulatory approach
- 7.5 Summary and conclusions
8 Educational perspective
- 8.2 AI impact on skills demand, learning and teaching
- 8.2.1 Direct AI impact on advanced skills demand
- --Impact of AI on learning
- --Impact of AI on teaching
- 8.3 AI skills and academic supply
- 8.4 Summary and conclusions
9. Economic perspective
- 9.1 Potential impact of AI on jobs
- 9.2 Potential impact of AI on growth
- 9.3 Potential impact of AI on inequality
- 9.4 Summary and conclusions
10. Cybersecurity perspective
- 10.1 Background: AI and cybersecurity
- 10.2 Applications of AI in cybersecurity
- 10.3 Deterrence and fight against crime
- 10.4 Robustness of AI algorithms against malicious action
- 10.5 Summary and conclusions
11. Computer processing and energy perspective
- 11.1 Introduction
- 11.2 Assessment of data centre (DC) energy consumption
- 11.3 Options to improve the energy efficiency of the increasing demand for HPC
- --CPU advancements: energy saving and parallelisation computing
- --Innovative and more efficient cooling systems/engineering solutions
- --Innovative infrastructures architectures and configurations
- 11.4 Summary and conclusions
12. Data perspective
- 12.1 The law and economics of access to data
- 12.2 The economic characteristics of data
- -- Applying economic reasoning to data access
- -- Policy intervention in the data market
- -- Summary
- 12.3 Towards a data strategy for public administrations
- -- Setting the scene
- -- About platforms and APIs
- -- From opening data to smart sharing
- -- Summary and conclusions
13 Societal resilience perspective
- 13.1 The need for resilience
- 13.2 Measuring resilience to AI
- 13.3 Lessons from the analysis of AI for understanding resilience
14. Summary and conclusions
This report presents a European view of Artificial Intelligence (AI) based on independent research and analysis by the European Commission Joint Research Centre to inform the debate at the European level.
We first introduce AI as a generic term that refers to any machine or algorithm that is capable
of observing its environment, learning, and based on the knowledge and experience gained, take intelligent actions or propose decisions. Autonomy of decision processes and interaction with other machines and humans are other dimensions that need to be considered.
Although many of the methodological developments in AI date back more than 50 years, the
reason why we now pay so much attention to AI in general and machine learning (ML) in particular is that the recent advances in computing power, availability of data, and new algorithms have led to major breakthroughs in the last six to seven years. The many applications of AI/ML have started to enter into our everyday lives, from machine translations, to image recognition and music generation, and are increasingly being exploited in industry, government and commerce (see Chapter 2).
It is likely that we are only at the beginning of this process because the development of ubiquitous sensor networks, the IoT, will increase exponentially the sensing capabilities of AI, the volumes of data on which to train the algorithms, and their reach in society
through decisions and actions. The opportunities are many, and in some cases not yet foreseen. There are also many challenges, however. Among them, current ML algorithms display some of the characteristics of a black box: we access the inputs and outputs but do not understand fully what happens in-between, and how certain outputs, including decisions and actions, are derived. This calls or a greater effort to understand their theory and to develop explainable and accountable algorithms. We also need strong evaluation frameworks that can assess not only the performance but also the quality of AI and build trust in this disruptive technology (Chapters 2, 6 and 10).
The overview of the global and European AI landscape shows that there is an intense competition on AI taking place world-wide with three main leaders: the USA, Europe, and China (Chapter 3). Each region has about one quarter of all key players in the AI field, including both research and industry, but has a distinctive different mix of players: while Europe is well balanced in the number of research and nonresearch players, the USA has approximately three times as many industrial/corporate players as research ones, and China has about six times as many research players as industrial ones. The strength of the corporate world in the USA is also indicated by its dominance in the number of start-ups (almost half
of the total worldwide) and venture capital (more than one third of the total). China, on the other hand, is making a strong effort to turn research research into patents, and accounts for almost 60 % of the world total. It has also put in place a strongly coordinated approach to AI, including government policy, industrial applications and research with the objective of becoming the world leader in AI by 2030. This is an ambitious but achievable target (Chapter 5). Europe is currently well positioned in the quality of its research production, with more than 30 % of all papers on AI published in top scientific journals, just behind the USA (33 %), and is considerably ahead of China. Key areas of strength in Europe on which to build upon are automated and connected vehicles and robotics.
We note that many European countries as well as the EC are developing strategies and programmes to guide the development of AI, with shared concerns over the need for an agreed ethical framework and applications that clearly benefit European society and uphold the European values enshrined in the Treaties (Chapter 4). The High-Level Expert Group
established in 2018 by the European Commission is elaborating a framework for subsequent developments linked to these values. We also note the high level of awareness that data is crucial to the development of AI, with policy documents at both national and European level putting strong emphasis on the need to share data better among all the stakeholders:
the public sector, industry, and the public. Finding the right way of doing it is challenging, as
analysed in Chapter 12 of the report.
After the overview and analysis of the AI global and European landscapes we discuss AI from multiple perspectives to add some depth and explore synergies. As we highlight at multiple stages in this report, a key characteristic of the European way to AI has to be a strong ethical framework. There is consensus on this principle, and we report (Chapter 4)
on many initiatives at both national and European levels to develop ethical guidelines to frame thdevelopment of AI. We review the main dimensions that such guidelines should consider and highlight the potential implications of AI at the level of individuals
and society. We conclude that to build and retain trust in AI we need a multi-layered approach that includes the critical engagement of civil society to discuss the values guiding and being embedded into AI, public debates in different fora to translate these values into strategies and guidelines, and responsible design practices that encode these values and guidelines into AI systems, so that they are ethical-by-design (Chapter 6).
In reviewing the European legal framework for AI, particularly with respect to the fundamental rights, data ownership, and intellectual property, we note the tensions between protecting rights of individuals and firms and encouraging innovation with trying to maximise openness and transparency. We conclude however that Europe is well placed to establish a distinctive form of AI that is ethically robust and protects the rights of individuals, firms, and society at large. For example, the General Data Protection
Regulation, opposed by many during preparation, is now perceived as a European asset and is inspiring similar approaches outside Europe. Extending this notion, we should consider the high standards of the European legal and regulatory landscape in a similar way to those in environmental quality which are an asset for Europeans and their future generations to build upon, not a barrier (Chapter 7).
From an educational perspective, we observe (Chapter 8) that AI has potential positive impacts on shortages of skills, learning, and teaching. Three crucial points stand out from the review: firstly, the need to understand better how the interaction with AI impacts human intelligence in cognitive capacities in both adults, and even more importantly, children. Secondly, we need to think beyond current needs and practices, and consider how AI is likely to change the relationship between education and work, and human development. Thirdly, we highlight possible risks related to AI in education, particularly privacy and ethical ones.
In this chapter, we also provide an initial overview and geographic distribution of the academic offer of study topics related to AI. This is relevant to the discussion in chapters 12 and 13 on possible strategies to start preparing society, and the most vulnerable regions, to the challenges that AI will bring.
The potential impact of AI on the labour market and inequality raises concerns in the media, research and public debates. We analyse the literature and evidence available to date on the potential impacts, both positive and negative, with respect to work, growth, and inequality. In relation to work, we find that neither theory nor evidence are very conclusive at the present time. AI could complement and enhance human activity, replace an increasing number
of routine tasks, or both. Studies measuring the share of jobs at high risk of automation exhibit a high variety in their findings depending on the definition and level of granularity at which tasks and occupations are defined (Chapter 9).
Another area of uncertainty is the extent to which AI has the potential to spur economic growth. When considered as a general-purpose technology, AI could spread across many jobs and industrial sectors, boosting productivity, and yielding strong positive growth. To the extent that ML generates new insights from data, it may also contribute to the production of new ideas and innovation itself. Economic growth models are starting to explore various scenarios but there is no empirical evidence yet that favours one or the other.
When it comes to inequality, we find that AI can affect unfavourably the distribution of income through many channels. The most discussed concern job polarisation, (i.e. increased demand for high-level, highly paid jobs on the one hand and low-level, poorly paid jobs on the other), reduction in job quality at the lower-skilled end, and also greater difficulty for lower-skilled workers to adjust to change and find new jobs, with longer periods of unemployment than those with higher qualifications and skills. These potentially negative
consequences on the labour market have a geographical footprint, as regions and subregions
that are already experiencing greater difficulties in terms of unemployment and low level of skills are likely to be the ones suffering most, if no action is taken. We clearly need to monitor closely and research the multiple impacts of AI on the economy in the coming years. We also need to consider a more proactive strategy to build the resilience of regions across Europe with a particular regard to the most vulnerable ones. We return to this in Chapters 12 and 13.
When it comes to cybersecurity, AI is a double-edged sword: it can be greatly beneficial to increase the security of devices, systems and applications, but can also empower those who seek to attack systems and networks and thus become an advanced tool in the arsenal for cyber-attacks. Moreover, the robustness of AI against malicious action itself becomes
an issue, posing the most immediate danger for the security of cyber-physical systems, in which AI will be increasingly deployed. There are lines of research focusing on understanding the specific vulnerabilities of AI and related attacks, and ways of increasing
AI robustness and interpretability and safety by design. We also need shared, large, high-quality datasets to train and test algorithms, and agreed frameworks to evaluate them (Chapter 10).
We discuss at several points in the report that the major progress we are seeing in the development of AI is linked to the rapid advances in computing and in data availability. With respect to computing, we are starting to see a paradigm shift due to the recognition that the increasing energy consumption of data centres and data transfers will become unsustainable in the era of the Internet of Things and 5G networks. Data traffic and processing loads are likely to be unprecedented when billions of additional devices will be connected to the internet, sensing the environment, and constantly sending and receiving data. With this in mind, we are seeing a trend towards more decentralised frameworks of edge and fog computing where processing is done closer to the sensors capturing and displaying data, including mobile phones and tablets. Industry has started moving in this direction but there is still a window of opportunity for European investment, regulatory frameworks, and standards to shape these developments so that they benefit Europe (Chapter11).
With respect to digital data, we review its key economic characteristics including economies of scale, of scope, and non-rivalry. The first two point to an increase in the concentration of data, and hence information and power, in the hands of a few actors in the internet economy, while non-rivalry creates potential tensions between opening access to the data so that society benefits the most, and restricting access so that the data holder benefits most. These tensions are reflected in the legal framework, as discussed earlier, and make the development of data strategies in an open and globalised environment a particular challenge. We argue, however, that if we apply the lessons of successful internet companies to the European public sector and develop ecosystems based around public platforms, it is possible to create large pools of shared data that are semantically well structured and labelled, and can fuel new AI applications in different domains. In this way we can open access to the data, develop
the market, serve the public, and enrich the data commons at the same time (Chapter 12).
We conclude our multi-perspective analysis of AI focusing on the concept of resilience which is useful to frame a European approach to AI addressing different phases: prevention, anticipation, preparation, but also adaptation and transformation to bounce back from the effects of a shock. Seen from this perspective, AI could not only be an engine for growth and change in Europe but also become an opportunity to bootstrap social and economic development in peripheral regions, leveraging the richness of European diversity and culture to the full (Chapter 13).
We put forward some elements of a possible approach centred on the network of Digital Innovation Hubs. There are already several hundred hubs which are expanding rapidly with a target of one for each region in Europe. The Communication on AI (EC, 2018a) foresees a specialised subset of these hubs on AI to facilitate access to technology and know-how to public administrations and firms, SMEs in particular, in combination with the planned
AI-on-demand-platform. We suggest putting these hubs at the centre of local ecosystems comprising public administrations, local enterprises, educational and training establishments, and civil society. These ecosystems could create local pools of shared data among the key actors so that the AI skills developed/upgraded locally could be put to good use in developing algorithms and solutions based on local data to address local needs. Each European region has its own specific mixture of priorities in relation to its environment, economy, demography, health, and so on. Therefore, this approach could provide an excellent opportunity to harness local creativity, culture, and knowledge of the territory to create socially relevant and people-centred AI and develop diverse and inclusive AI systems (Chapters 12 and 13).
From this multi-disciplinary analysis, we draw the following conclusions:
We are only at the beginning of a rapid period of transformation of our economy and society due to the convergence of many digital technologies. Artificial Intelligence (AI) is central to this change and offers major opportunities to improve our lives. The recent developments in AI are the result of increased processing power, improvements in algorithms and the exponential growth in the volume and variety of digital data. Many applications of AI have started entering into our every-day lives, from machine translations, to image recognition, and music generation, and are increasingly deployed in industry, government, and commerce. Connected and autonomous vehicles, and AI-supported medical diagnostics are areas of application that will soon be commonplace.
There is strong global competition on AI among the US, China, and Europe. The US leads for now but China is catching up fast and aims to lead by 2030. For the EU, it is not so much a question of winning or losing a race but of finding the way of embracing the opportunities offered by AI in a way that is human-centred, ethical, secure, and true to our core values.
The EU Member States and the European Commission are developing coordinated national and European strategies, recognising that only together we can succeed. We can build on our areas of strength including excellent research, leadership in some industrial sectors like automotive and robotics, a solid legal and regulatory framework, and very rich cultural diversity also at regional and sub-regional levels.
It is generally recognised that AI can flourish only if supported by a robust computing infrastructure and good quality data:
- With respect to computing, we identified a window of opportunity for Europe to invest in the emerging new paradigm of computing distributed towards the edges of the network, in addition to centralised facilities. This will support also the future deployment of 5G and the Internet of Things.
- With respect to data, we argue in favour of learning from successful Internet companies, opening access to data and developing interactivity with the users rather than just broadcasting data. In this way, we can develop ecosystems of public administrations, firms, and civil society enriching the data to make it fit for AI applications responding to European needs.
We should embrace the opportunities afforded by AI but not uncritically. The black box characteristics of most leading AI techniques make them opaque even to specialists. AI systems are currently limited to narrow and well-defined tasks, and their technologies inherit imperfections from their human creators, such as the well-recognised bias effect present in data. We should challenge the shortcomings of AI and work towards strong evaluation strategies, transparent and reliable systems, and good human-AI interactions.
Ethical and secure-by-design algorithms are crucial to build trust in this disruptive technology, but we also need a broader engagement of civil society on the values to be embedded in AI and the directions for future development.
This social engagement should be part of the effort to strengthen our resilience at all levels from local, to national and European, across institutions, industry and civil society. Developing local ecosystems of skills, computing, data, and applications can foster the engagement of local communities, respond to their needs, harness local creativity and knowledge, and build a human-centred, diverse, and socially driven AI.
We still know very little about how AI will impact the way we think, make decisions, relate to each other, and how it will affect our jobs. This uncertainty can be a source of concern but is also a sign of opportunity. The future is not yet written. We can shape it based on our collective vision of what future we would like to have. But we need to act together and act fast.