21 March 2019

Tackling Europe’s gap in digital and AI

By Jacques Bughin

Europe’s average digital gap with the world’s leaders is now being compounded by an emerging gap in artificial intelligence.


On many metrics, the European economy and its businesses have been grappling for years to capture the full potential of current and previous generations of digital tools. It is now more than time to double down on Europe’s efforts to succeed in digital transformation, especially when a new set of digital technologies such as artificial intelligence (AI), are becoming more technically pervasive.
Stay current on your favorite topics

On average, Europe’s digital gap with the world’s leaders is now being compounded by an emerging gap with the world’s leaders in its development and corporate use of AI technologies. Without faster and more comprehensive engagement in AI, that gap could widen, especially for those European countries with relatively low AI-readiness.

The potential to deliver on AI and catch up against the most AI-ready countries such as the United States and emerging leaders like China are large. If Europe on average develops and diffuses AI according to its current assets and digital position relative to the world, it could add some €2.7 trillion, or 20 percent, to its combined economic output by 2030. If Europe were to catch up with the US AI frontier, a total of €3.6 trillion could be added to collective GDP in this period.


One positive point to note is that Europe may not need to compete head to head but rather in areas where it has an edge (such as in business-to-business [B2B] and advanced robotics) and continue to scale up one of the world’s largest bases of technology developers into a more connected Europe-wide web of AI-based innovation hubs.

In a new discussion paper, Notes from the AI frontier: Tackling Europe’s gap in digital and AI(PDF–623KB), the McKinsey Global Institute (MGI) blended the findings of authoritative secondary research sources with three primary independent global surveys at the corporate and sector levels conducted in 2017 and 2018 to better gauge how firms anticipate the way AI might unfold in Europe. The research also updates MGI’s comprehensive global model of the diffusion of AI developed for early research on AI for the EU-28, in particular integrating a perspective on the development of AI startup ecosystems in Europe.

Among the major findings are the following:

Europe is adding an AI gap to its digital gap

MGI’s 2016 research found that European countries were capturing only 12 percent of their full digital potential (defined as weighted deployment of digital assets, labor, and practices across all sectors, compared with the most digitized sector)—two-thirds of the captured potential in the United States.

Using the most recent data from McKinsey’s digital survey in 2017, the same gap remains. Europe is not standing still, but the pace of AI diffusion and investment remains limited. Although Europe’s GDP is comparable with that of the United States and just ahead of China’s, the digital portion of Europe’s ICT sector today accounts for around 1.7 percent of GDP, lower than the share in China at 2.1 percent and only half the 3.3 percent share in the United States. While large Western European companies are continuing to expand their use of early digital technologies, the share of fully digitized companies increased by less than 10 percent a year between 2010 and 2016.

VideoHow can the EU scale up artificial intelligence?

Jacques Bughin, a director at the McKinsey Global Institute, explains the three elements necessary to scale up artificial intelligence and where Europe stands.

Europe has some solid assets to bring into play in the next wave of AI. For instance, it has close to six million professional developers—more than in the United States. Yet Europe’s disadvantage in digital diffusion seems likely to spill over into AI, where a new gap is appearing. Early digital companies have been the first to develop strong positions in AI, yet only two European companies are in the worldwide digital top 30, and Europe is home to only 10 percent of the world’s digital unicorns. Europe has about 25 percent of AI startups, in line with its size in the world economy, but its early-stage investment in AI lags behind that of the United States and China.

Further, with the exception of smart robotics, Europe is not ahead of the United States in AI diffusion, and less than half of European firms have adopted one AI technology, with a majority of those still in the pilot stage. AI initiatives remain fragmented in Europe, and investment in AI is nothing like the size of that in the United States or China. Europe attracted only 11 percent of global venture capital and corporate funding in 2016, with 50 percent of total funds devoted to US companies and the balance going to Asia (mostly China). That share was about the same in 2018. Only four European companies are in the top 100 global AI startups.

Available data on diffusion are scarce, but our blend of survey research demonstrates that European companies may lag behind their US counterparts in their adoption of big data architecture and of the advanced machine learning techniques that are the foundations of AI—with 12 percent less use than in the United States. A possible gap may exist between Europe and the United States on the use of AI tools such as smart workflows, cognitive agents, and language processing—a 16 percent gap to date. Moreover, European AI is yet to be deployed broadly in enterprises rather than in one or only a few functions.

Only 5 percent of European AI adopters (compared with about 8 percent in the United States) are using these tools in about 90 percent of their entire organizations. Seven of ten companies, however, are capturing 10 percent of full potential use. In the most advanced industry—high-tech—93 percent of adopters are capturing AI for 10 percent of its potential use, but still only 17 percent of European companies (compared with about 22 percent in the United States) are using AI technologies at 75 percent of potential. At the other extreme, only 2 percent of European firms in healthcare systems and services are using those technologies at 80 percent of potential.

AI may scale up in a fast-paced game of competition, innovation, and new skills acquisition

Our analysis of three surveys suggests that there are three channels that will determine the extent of the productivity boost that comes from AI: competition, innovation, and new skills.

European companies perceive AI to be a competitive play. When European companies that have not yet invested in AI were asked whether they see some competitive risks from both AI-native firms and early AI adopters among incumbents, nonadopters perceived there to be equal risk from both types of companies. For example, a majority (53 to 57 percent) of nonadopters believed that both can engage in aggressive taking of market share from competitors. In our survey, the primary objective for 15 percent of European companies investing in AI (slightly below the share of Asian and North American firms, at just over 20 percent) is taking market share from competitors. Our survey clearly shows that digitally savvy European companies are 15 to 25 percent more likely to use AI.

Notes from the AI frontier: Tackling Europe’s gap in digital and AI

Download the discussion paper on which this summary is based (PDF–622KB).

AI’s potential to deliver revenue growth through innovations rather than efficiency alone is motivating European adopters. Our various surveys consistently find, however, that companies are equally—if not more frequently—motivated by the pursuit of capital productivity and the efficiency of non-labor inputs. About 30 percent of European adopters report that they are using AI with an eye to revenue expansion, whether through extending into new markets or gaining market share. Companies with less experience in AI tend to focus on its ability to help cut costs, but the more that companies use and become familiar with AI, the more potential for growth they see in it.

Why are some companies absorbing AI technologies while most others are not? Among the factors that stand out are their existing digital tools and capabilities and whether their workforce has the right skills to interact with AI and machines. Only 23 percent of European firms report that AI diffusion is independent of both previous digital technologies and the capabilities required to operate with those digital technologies; 64 percent report that AI adoption must be tied to digital capabilities, and 58 percent to digital tools. Our surveys report that the two biggest barriers to AI adoption in European companies are linked to having the right workforce in place. The first barrier relates to the ability to use ICT tools in work. The second barrier relates to companies’ need for skills to provide new AI applications and services, such as AI coding and analytic expertise.

These drivers of adoption have major implications for European business and society:

The competitive edge associated with AI and the fear of being disrupted may lead to a competitive race—such a race has already begun between China and the United States in the internet sector. The various surveys we used lead us to estimate that competition among firms may account for 50 percent of European corporates’ decision to adopt AI by 2030.

AI will be used for new business models, products and services, and skills, which suggests a large transformation of most jobs, including an emerging talent war for AI and digital skills coupled with creative skills. Such a talent war is already visible today.

AI may have substantial positive implications for economic growth and productivity but may feature winner-take-most dynamics in many industries, as companies are likely to adopt at very different paces and therefore have different abilities to take advantage of the opportunities afforded by AI. Our average case estimate suggests that the 10 percent of European companies that are the most extensive users of AI to date are likely to grow three times faster than the average firm over the next 15 years. That dispersion in productivity gains in favor of front-runners has similar characteristics to the recent phenomenon of superstar firms.

AI could give EU economies a strong boost

If Europe develops and diffuses AI according to its current assets and digital position relative to the world, it could add some €2.7 trillion, or 20 percent, to its combined economy output, resulting in 1.4 percent compound annual growth through 2030 (Exhibit 1). Such an impact would be roughly double that of other general-purpose technologies adopted by developed countries in the past.
Exhibit 1

If Europe further improves on its assets and competences sufficiently to catch up with the United States’ AI frontier, the potential could be even higher. GDP growth could accelerate adding an extra €900 billion to GDP and bringing the total potential AI boost to €3.6 trillion by 2030.

In our analysis, we examine three macroeconomic enablers—automation potential, investment capacity, and connectedness—and four microeconomic enablers: digital legacy, innovation foundation, human capital, and the maturity of AI ecosystems. In general, Europe fares well with regard to its automation potential and in the stock of cognitive skills, but has not, on average, been able to increase its innovative capacity, and faces challenges in developing a large AI startup ecosystem.

Europe may achieve a significant productivity boost through AI, without sacrificing employment in the long term. Throughout history, technology has eliminated some types of jobs, but it also always created new ones. It is impossible to predict with any precision all of the jobs that are likely to be created through AI, but we contend that in the EU-28, on average, AI could enable the creation of as many new jobs as jobs that are changed, especially if Europe develops innovative new products and new demand. More innovation, fluidity in job reallocation, and internalization of AI gains (mostly by taking major positions in the AI supply chain) within Europe is likely to determine the fate of job development in the region. Powerful development of AI may be the best hedge and may even be the catalyst for new jobs in Europe in the future.

AI performance is likely to vary among EU member states

Europe’s average ability to capture the full potential of AI masks a significant disparity among countries and sectors.

The total effect of AI on GDP and employment growth should depend on whether a set of core AI enablers are in place—and are nurtured.

We collected a set of indicators by country to gauge how they stand on the key enablers and aggregated them into an AI Readiness Index. Index scores are not pure averages but are based on weighting each enabler according to its relative importance for boosting the economic growth of each country (Exhibit 2). The following are some of the findings:

The most advanced Northern European countries and the Anglo-Saxon countries lead in Europe, ahead of China (and just behind the United States).

The United States leads the index, driven by a strong AI ecosystem, positive ICT connectedness, and strong innovation capabilities. China is notably able to reinvest lots of its gains into the economies and is already deploying AI ecosystems. In general, automation potential is lower in China than in Europe because of lower incentive to arbitrage salaries.

A clear gap in AI readiness exists, with Southern and Eastern Europe lagging. The main driver of differences between the most AI-ready and the least reflects slower AI adoption in less ready countries that limit the potential benefits of the competitive race to AI, lower skills with which to reap the benefits from AI, and a lower share of innovative firms leveraging AI.
European countries have very different strengths and weaknesses on the enablers. For instance, Ireland tops the index on ICT connectedness, Finland on human capital, and the United Kingdom on innovation. The dispersion of strengths indicates that countries can borrow best practice from each other to create a more favorable and more enabling environment for AI.
Exhibit 2
Europe should consider prioritizing action in five areas to accelerate its path to AI

At the very least, Europe needs to develop its journey toward AI based on the enablers it already has and, therefore, its current readiness for AI. That may not be sufficient for a large number of European countries that may be at risk of exclusive growth, however. A more ambitious aim would be for Europe to try to close the gap with leaders such as the United States and China. Remember that these two world AI leaders may be forging ahead aggressively, and Europe runs a distinct risk of falling further behind in the race to AI and facing more competition for capturing growth and employment.

If Europe fails to accelerate its adoption and diffusion of AI, it is likely to achieve only minimal productivity growth gains through the use of AI but may also fail to catch up with the United States and China. We foresee five priorities on which Europe should focus:
Europe needs to continue developing a vibrant ecosystem of deep tech and AI startup firms that will use AI to create new business models

Europe’s incumbent firms need to accelerate their digital transformations and embrace innovating with AI

Progress on the digital single market is continuing but still incomplete

To capture the opportunity, companies need to build the right talent and skills
Think boldly about how to guide societies through the potential disruption

Europe still suffers from a digital gap. Given that digital technologies are the bedrock of diffusion of AI technologies, the risk is that Europe could fall further behind the world’s leaders on AI technologies and miss out on a significant source of potential new economic dynamism. We know that AI shares winner-takes-most characteristics with the previous wave of digital technologies, and that therefore there is an urgent imperative for Europe to build on current strengths and pockets of best practice—and up its game. In short, Europe needs more AI, different AI, and all of it more quickly. This paper offers some brief thoughts on a road map of priorities that need to be in the mix.

No comments: