Artificial intelligence (AI) is at the cutting edge of innovation and stands out as a transformational technology of our digital age.
AI is generating a lot of interest But the reality is that the field has finally begun to deliver on the promises it first made back in the 1950s.
Machine Learning systems as one of the most crucial areas of AI have shown superhuman performance in areas ranging from recognising object in pictures, diagnosing diseases, detecting fraud, making customised recommendations to shoppers, and playing chess (with outstanding success).
AI’s practical application throughout the economy is growing apace and the new generation of applications is based on a digital foundation.
Early adopters are already gaining competitive advantages, and the gap with the laggards is set to grow. Perhaps most important of all, the world is now generating vast quantities of the fuel that powers AI : data.
Globally, there has been a huge increase in data available. The digitalisation of information, the expansion of the Internet and mobile networks and the progressive deployment of the Internet of Things (IoT) have created a massive amount of data.
In addition, data storage costs have decreased dramatically and the use of cloud environments and big data technologies have facilitated the access to this data.
The availability of computer power has increased the speed and accuracy of AI technologies.
Distributed computing has become popular and is available for researchers at limited cost. Moreover, there have been improvements in the hardware associated with AI, using new parallel architectures that are increasing the speed five to six fold every year.
AI is intrinsically linked to digitalisation, as it is embedded in numerous technology applications.
For example, the fundamental aspect of machine learning is that it learns from itself, from exploring and discovering pattern and logic in the data. Thus, obviously, data is key and the more data the better implying that scale is key.
The question then becomes, how to reach scale? By having a large and well-developed digital economy, where data constantly can flow freely and in massive quantities.
Both the US and China can benefit from highly developed digital industries, which a more digitally fragmented Europe cannot.
The large tech giants all have the necessary ingredients to succeed in an AI driven future: access to data, the financial means to invest in state- of-the-art technology and buy necessary data, they also have sophisticated search engines and they have control over the IT infrastructure and thereby also the ability to attract skilled human resources.
However, luckily, there are tremendous AI application opportunities in more traditional industries in industries where Europe in fact is leading, such as the automotive sector, energy or industrial automation.
However, these sectors are underrepresented in AI in Europe. Since they are large and EU companies are at the frontier, we could expect technological development within AI to also be at the frontier.
Problem is key sectors in Europe have not yet embraced AI meaning that many of the true engines of the European economy have not yet embraced AI technologies.
Why not? And how to change that? Here comes the financing aspect into the picture. External AI investment is growing fast estimated to have been growing by three fold over three years as well as internal investment, dominated by cash-rich digital native companies, which are highly concentrated in a few technology hubs in the US and China.
Low investment in AI is a challenge for Europe: the US invests about 5-6 times more, and Asia triple the amount of the EU giving rise to a substantial investment gap.
This gap, given the overall digital dominance of these two regions, is set to grow if no actions urgently are being undertaken.
Besides, with Brexit looming, the lagging performance of the EU will only be reinforced as up to a third of all European AI activities are estimated to be taken place in London.
In line with other new emerging technologies and RDI in general but RDI at the frontier/cutting-edge research in particular investment in AI is associated with high risk.
AI involves both a commercial risk: as the markets may not be mature enough and concrete applications may not be viable; and a technological risk since it involves high uncertainty whether the R&D will lead to a successful outcome (high risk of failure).
Who are willing to take on higher risk? Equity investors, rather than banks.
This points to a problem for Europe’s bank based financial system. Another problem, even if equity investor would be willing to take the risk, they typically have a short-term investment-return horizon which may not be aligned with the long-term development cycle of AI RDI activities.
This calls for a longer-term view on return on AI investments, in particular so called patient capital.
As the EU bank, the EIB provides long-term finance for sound, sustainable investment projects in support of EU policy goals in Europe and beyond.
At the EIB, we support companies throughout their business life cycle. What we believe is a particularly crucial phase in the case of AI is the scale-up phase.
We stand committed to support the growth of promising companies, as well as experimental research.
We understand that we have to create a truly European innovation ecosystem where EU-based start-ups can grow and reach a sufficient scale in their home continent.
We support the creation of an “AI ecosystem”, with clusters of entrepreneurs, financiers, and AI users. AI technologies are seen by the Bank as an important and growing part of the on-going digitalisation of the economy.
Early adoption of AI capabilities will not only increase the operational efficiency of businesses, but also facilitate identification of new opportunities, new products, new customers and new channels, creating competitive advantages across many different industries.
On the demand side, the use of AI has the potential to becoming a key factor for the competitiveness of European companies in the world market.
Companies in various sectors, such as finance, retail or manufacturing, could benefit from EIB loans in order to accomplish the required investments to digitalise their businesses and subsequently embed AI components in the decision making processes.
On the supply side, the EIB could play a role by supporting traditional software companies in the development of new AI related capabilities as well as being a fundamental source of long-term financing for SMEs that are developing AI technologies, ensuring the successful transition from R&D to commercialisation.
AI technology is cutting-edge, hence proof of concepts and cost benefit analyses are not easily built, which limits access to finance from traditional financial institutions and provides an excellent opportunity to take advantage of EIB objectives to address such market failures.
The Bank stands ready to help financing the success of Europe’s technology-driven future and support investments in development and deployment of AI, crucial for EU’s competitiveness.
We combine proven financial expertise with the capacity to take on additional risk in order to explore new fields of innovation and help companies in the critical growth-stage.
And we are in it for the long haul. EIB financing will reach the front lines of technological development, so that those sitting in the driving seat will keep on moving forwards, and those in the back seat keep on following.
Now is about time to accelerate the EU innovation machine.