Can Europe catch up on A.I.?

If you were asked to bet on who will lead the world in Artificial Intelligence by 2030, which country would you pick? China, the United States, India? It is difficult to imagine that one would bet on Europe. If this is the case for you, this article might change your mind.

The article aims to show the gap between Europe and the rest of the world in terms of investments, research and development concerning A.I. It will also estimate the best approach to the question “Can Europe catch up on A.I.?”.

In order to make our quick analysis, we are going to follow the path of a Deutsche Bank Research paper published at the beginning of 2020, titled “(How) will the EU become an A.I. superstar?” and two McKinsey reports, “Notes from the AI frontier: Tackling Europe’s gap in digital and AI” and “How nine digital frontrunners can lead on AI in Europe”.

One of the most important preconditions for the spread of A.I. is the advancement and adoption of digital technologies. Europe’s digital gap, which is visible in exhibit 1 of McKinsey report (showing the percentage of digital ICT on GDP for USA, CH, EU), risks to critically affect its chances of playing a significant role on A.I. Source: McKinsey

Similarly, Exhibit 10 of the McKinsey report shows the dependence of AI development on previous adoption of digital technologies generalized to 430 European companies. Only 23% of these companies affirm to be independent of both technology tools and capabilities. Source: McKinsey

Source: Deutsche Bank Research

Source: Deutsche Bank Research.

Graph 3 (on the left) of Deutsche Bank’s report, evidently depicts the predominance of the US, China and Israel in the numbers of recorded A.I. startups, closely followed by former-EU-member the United Kingdom, Canada and Japan. France (the first European country on the list) can be seen in seventh place. Furthemore, chart 4 (on the right) illustrates that only five of the top 100 global A.I. startups (those considered most promising by CB Insights) in 2020 were from EU27. The majority of them originated in the US, the UK and China.

Analyzing the amount spent on R&D on software and computer services (good indicators of related A.I. Research), we immediately notice a persistent gap between Europe and the US. If we take the global R&D spending into account we notice that the EU accounts for only 4.7% of it while China and the US account for 16.7% and 26.8% respectively. (Graph 14 from the Deutsche Bank report).

Another important factor influencing R&D on A.I. and new technologies is the dominance in big tech companies in the US and China, compared to the EU. These companies focus on concentrating A.I. talent and research capabilities as they try to speed up the transition process towards A.I., as a means of establishing a dominant role in the new market.

Source: Deutsche Bank Research

It is also important to address the difference in terms of private investments in A.I. between Europe, the US and China. We notice that only 8% of total private AI investments take place in Europe, compared to 46% in the US and 36% in China. (Exhibit 7 from the McKinsey report). Source: McKinsey

It is evident that Europe experiences a significant deficit in advanced technology patents, such as Artificial Intelligence, compared to the US. (Exhibit 19, McKinsey report).

A lot has been discussed at this point. However, comparisons previously made ought to take into consideration the fact that Europe’s market is indeed comparable to that of the US and China in size. Nevertheless, Europe positions far behind both China and the US in the race for A.I. dominance. Source: McKinsey

One explanation for this phenomenon is the fact that Europe’s market is very fragmented, as shown in Chart 10 of Deutsche Bank report. An important component that many big tech companies have in common is the fact that they were either created in a big and integrated domestic market (e.g. US) or in a big and protected market (e.g. China). Therefore, solving Europe’s fragmentation issue could possibly be one of the most effective ways of catching up with the US and China.

Source: Deutsche Bank Research

There is, however, one aspect in which Europe has demonstrated to be a front-runner: A.I. regulation. This means that Europe has been more consistent and efficient than the US and China on addressing the problem of regulating digital data by adopting policies such as the General Data Protection Regulation (GDPR).

Europe’s strict approach towards digital technologies and A.I., though being fundamental, is another element that strongly contributed in broadening the gap with the US and China. This approach has yet to be addressed by countries currently adopting less rigorous policies.

Therefore, it is a matter of finding the right balance between protecting user rights and stimulating the development and commercialization of AI.

Furthermore, in February 2020, the European Commission published the “White Paper on Artificial Intelligence - A European approach to excellence and trust”, which highlighted the necessity for a solid European approach to be built on the European strategy for AI presented in April 2018.

As stated, “To address the opportunities and challenges of AI, the EU must act as one and define its own way, based on European values, to promote the development and deployment of AI.”

In the white paper, the European Commission announces the aim to pursue both a regulatory and an investment-oriented approach. It also underlines the necessity to set up a common European approach to AI and avoid market fragmentation.

The objective is to “become a global leader in innovation in the data economy and its applications”.

In the conclusion of its report, Deutsche Bank points out three priority measures that Europe must adopt in order to become an AI superpower.

• Gaining A.I. competitiveness. It is essential that Europe acts rapidly and boldly (through substantial public and private funding) in order to avoid the threat of falling far behind in the global competition for AI leadership.

• Setting the A.I. playing field. The EU must adjust and regularly update the A.I. regulation, without the need to find wholesale solutions right from the start.

• A human-centric approach to A.I. It is essential that A.I. is developed in line with fundamental values and benefits all citizens without creating further economic inequality.

To sum up, Europe entered the race on A.I. with a substantial disadvantage compared to its competitors. Through time, though, it has shown to understand the importance of the challenge by adopting new and common policies, especially in the last years.

Covering the gap with the US and China is challenging, but we must keep in mind that the odds of Europe being THE A.I. superpower were close to zero to begin with. However, there remains the possibility for Europe to carve out a competitive (and hopefully determinant) role in Artificial Intelligence and related markets in the years to come.

Cover picture from Pixabay