Tackling Europe’s gap in digital and AI
By Jacques Bughin
Europe should have been in the vanguard of the artificial intelligence (AI) revolution starting to transform business. After all, the world’s first autonomous car was a German Mercedes, 35 percent of the deep-learning models running in the world were invented in Europe, and Europe is home to more software developers than the United States. However, Europe is falling behind in the AI race — just as it did in the race to digital.
New McKinsey Global Institute (MGI) research finds that, on average, the 28 member states of the European Union (EU-28) could potentially add some €2.7 trillion, or 19 percent, to GDP by 2030, if they absorb AI according to their current competencies and assets. If they were to accelerate and catch up with the United States as a group, the potential could be as high as €3.6 trillion, while the risk of not joining this race may actually lead, in the extreme, to a pressure on productivity and growth.
Despite the depth of its leading-edge talent, Europe may start from a weak position in its bid to capture those growth benefits. Digital technologies are the bedrock of AI, but Europe has yet to close its digital gap with the United States. It still stands about 35 percent behind the world’s leader. Large Western European companies are expanding their use of early digital technologies, but the share of fully digitized companies increased by less than 10 percent a year between 2010 and 2016.
Europe’s investment into AI supply chain is far lower than in the United States and China, the two countries leading the field. Only 11 percent of global venture capital and corporate funding of AI went to Europe in 2016, with half going to the United States and most of the rest to China. Two years later — in 2018 — Europe may not have increased its share, according to data from CB Insights. While Europe witnesses a growing number of inceptions of AI start-ups, in line with the US, there is still a risk they do not scale. Currently, only four European companies are represented in the top 100 AI start-ups.
We also find that European companies’ adoption of big data and advanced machine learning is about 12 percent behind that of their US counterparts. Their use of AI tools such as smart workflows, cognitive agents, and language processing lags behind by about 16 percent. Overall, we find that Europe may, on average, be about 30 percent less AI-ready than the United States. The exception is smart robotics which Europe, -especially due to Scandinavia and Germany, has historically been rather eager to use in manufacturing.
Some European countries are more ready for the AI revolution than others. We looked at key enablers of AI supply chain deployment and AI adoption such as the extent of digitization, the strength of a country’s innovation foundations, and its advanced cognitive skills. We then developed an AI Readiness Index based on the relative weight those different enablers may have on boosting AI productivity growth. The United States tops the index. Within Europe, the United Kingdom, Sweden, Finland, and Ireland are in the top quarter of AI-ready countries. Croatia, Greece, Hungary, Poland, and Romania are in the bottom quarter. European countries have very different strengths and weaknesses. For instance, Ireland tops the index on ICT connectedness, Finland on human capital, and the United Kingdom on innovation.
We know that the adoption of digital and AI is a competitive race, and that those that move early win most of the benefits. If Europe wants to commit to AI, catching up is therefore urgent. But nobody should underestimate the challenges that lie ahead. There would be large upfront costs to absorb before the value of AI appears. We reckon around 40 percent of the eventual value or around €0.8 trillion will need to be invested upfront to manage the transition — helping people through short-term unemployment and, critically, supporting then to learn the skills they will need to work in the AI age.
Europe is already short of the advanced skills it would need to capture the growth potential of AI while preserving jobs. We find that 26 percent of companies that adopt AI are concerned that they will not be able to secure the skills they need. But if Europe invests and manages the transition to AI effectively — and crucially if European companies adopt AI to innovate rather than simply to make efficiency gains — as many new jobs could be created as AI changes or displaces.
Effort would be required on a broad front beyond skills and bold thinking about how to manage the strains and cost of the transition to AI. Europe would need to be aggressive and creative about supporting even more vibrant deep tech and AI startups. Consider that Stockholm is one of the most prolific technology hubs in the world in per capita terms; as early as 2013, Sweden’s capital had 250 deep tech firms, of which nearly one-third were AI-based startups. Yet Sweden still invested only half what the United States poured into digital startups from 2015 to 2017 — although double what Finland and the United Kingdom spent.
European companies would need to move faster not only on their digital transformations but into AI — and innovation powered by AI. Less than 5 percent of European companies have radically redesigned business models using AI, compared with more than 8 percent in the United States.
Public policy can help. Strategic AI plans are sprouting up across the continent — Finland, France, Germany, the United Kingdom all have them — that signal a strong commitment to embracing AI. However, these efforts are fragmented. The trick will be coordination, and one of the most powerful contributions that European policy makers could make would be to complete the digital single market.
Society has many concerns about AI including cybersecurity, data security, and job security, all of which need to be addressed. But the economic case for AI is compelling, and there is a danger could fall further behind the world’s leaders and miss out on a significant source of potential new economic dynamism.
Read more in MGI’s new research, Tackling Europe’s gap in digital and AI.
Jacques Bughin is a director of the McKinsey Global Institute (MGI) and a senior Partner in McKinsey’s Brussels office. This article was originally published on LinkedIn.
