Looking at the data from the Digital Wallonia 2020 platform, we can see that Artificial Intelligence (AI) is a strategic advanced technology for Wallonia. About the AI ecosystem: digital solution providers, research stakeholders and organisations offering support.
AI: multiple technologies and applications
Artificial intelligence (AI) is “the set of theories and techniques used to create machines that are capable of simulating intelligence”. It is a set of concepts and technologies rather than an autonomous discipline.
To cite the online course, Elements of AI: “...the term AI means different things to different people. For some, AI refers to artificial life forms that can surpass human intelligence, while for others, any technology that processes data can be referred to as AI...”
AI comes in many forms and offers multiple opportunities for its users in terms of applications. It can also help us make better use of other technologies, such as robotics. According to Forrester (2016-2017), the concept of AI covers the following (non-exhaustive list):
- Automatic text generation (customer services, generating reports and business intelligence insights).
- Automatic speech recognition (interactive voice response systems and mobile applications).
- Virtual agents (customer service and smart home management).
- Machine learning platforms (enterprise applications, mainly involving prediction or classification).
- AI-optimised hardware (efficient running of AI-oriented computational jobs).
- Decision management (enterprise applications, assisting in or performing automated decision-making).
- Deep learning (recognising patterns and classifying applications, supported by big data).
- Biometric recognition (mainly used in market research and security).
- Robotic process automation (automating human action, to carry out certain tasks efficiently).
- Text analysis and automatic natural language processing (fraud detection, security, automated assistants and applications for mining unstructured data).
In terms of social applications, virtually all aspects of human life have the potential to be affected.
Artificial intelligence in Europe
AI is seen as a top priority by the European Union within the context of its Digital Europe Policy (DEP) for the coming years, together with cybersecurity and HPC (high-performance computing).
The European Commission’s standout initiatives
- The publication of the first guidelines for an EU AI strategy in April 2018, stressing the need to :Put Europe at the forefront of AI technological developments and encourage the adoption of AI by businesses and public authorities.Prepare for the socio-economic changes brought about by AI.Establish an appropriate ethical and legislative framework.
- The White Paper on AI, in February 2020, outlining :Measures to boost investment, research and cooperation among Member States in the development and deployment of AI.Proposals for a European regulatory framework defining the legal requirements that would apply to the different stakeholders.
- Report entitled “Safety and liability implications of AI, the Internet of Things and Robotics”, released in February 2020.
- A public consultation, in June 2020, dedicated, among other things, to reviewing the White Paper.
- The creation of the European AI Alliance. This is a multi-stakeholder forum that analyses the growth of AI and its socio-economic implications. Its main goal is to help define the Commission's AI strategy and action plan. In particular, it will put together recommendations for ethical and legislative issues and identify the challenges and opportunities introduced by AI in the medium and long term.
- The launch of a group of leading AI experts. Uniting 52 experts from business, universities and civil society, this group has produced four deliverables over the course of two years dedicated to the ethical aspects of introducing AI and recommendations for investment in healthcare, industry and the public sector.
Lastly, a coordinated AI Plan for Europe is due to be released as early as the first quarter of 2021.
These initiatives should not blind us to the significant differences between Member States when it comes to support for and adoption of AI. For example, in 2018, Roland Berger grouped European countries into three categories: leaders, rising stars and “followers”. Belgium was (and probably still is) in the third category.
According to Asgard (2017), the same trends can be found when it comes to support hubs for AI startups.
Initiatives in Belgium
The situation has obviously evolved since then, with a number of countries launching ambitious AI research and funding programmes. In Belgium, this has resulted in :
- the federal initiative AI4Belgium,
- a 30 million Euro public/private investment plan in Flanders (March 2019),
- the DigitalWallonia4.ai programme (July 2019). This now encompasses all the initiatives relating to AI in Wallonia, including the Trail consortium, for example.
There is indeed a sense of urgency, as Europe’s seven “AI champions” are currently non-European companies.
AI: a major challenge for Wallonia
DigitalWallonia4.ai is committed to speeding up the adoption of artificial intelligence (AI) in Wallonia and consolidate its ecosystem. The strategy can be broken down into four key areas, which are in turn made up of a number of initiatives:
- Focus area 1 : Society and AI;
- Focus area 2 : Businesses and AI;
- Focus area 3 : Training and AI;
- Focus area 4 : Partnerships, Innovation, Research and AI.
DigitalWallonia4.ai is supported by the AI Network, Agoria, the Infopôle ICT Cluster and the Agence du Numérique.
DW4AI also relies on support from the network of Digital Wallonia International Hubs set up in partnership with AWEX for international intelligence gathering.
The AI ecosystem in Wallonia
Based on data from the Digital Wallonia platform (in November 2020), Artificial Intelligence is the second most important cutting-edge digital technology in Wallonia, behind the Internet of Things and ahead of Virtual/Augmented Reality and even robotics and automation.
AI therefore offers a huge amount of potential for innovation, for developing a digital offering and cutting-edge expertise, as well as when it comes the digital transformation of the region’s businesses.
So Wallonia is currently home to :
- More than 130 businesses in the digital sector offering products and services related to AI.
- Around fifty research stakeholders developing basic and applied research projects in this field, or supporting digital companies specialising in AI.
- Around ten Digital Wallonia partners supporting or guiding digital companies when it comes to developing AI solutions.
AI: a cutting-edge technology that is popular with (young) digital stakeholders
Private AI stakeholders can be broken down into 70 companies and SMEs, 60 startups and 6 spinoffs. Together, they represent 6% of Wallonia’s digital businesses on the Digital Wallonia platform. However, the proportion of startups is significantly higher than that seen for other categories of products and services.
Within the “Advanced” subcategory of digital products and services, 30% of businesses have an AI offering and, if we just look at startups, this proportion rises to over 40%. This should be no surprise: AI is still a young technology in the Walloon economic landscape, so it is only natural that there is a large number of recently created businesses among those offering AI solutions.
There is also some encouraging news when it comes to research. Whether they are accredited research centres, research units in “haute écoles” orin universities, around 50 are working in AI, so over half of the research stakeholders listed and more than 2/3 of those involved in the “Advanced” sub-category.
While this is only a quantitative analysis and ignores the total value of investments or the number of people actually dedicated to AI within each centre, it could be argued that research in this field is very healthy in Wallonia and, at the very least, has real potential. Once again, it is regrettable that there is still a significant discrepancy between the structure of the digital economic sector and the world of research.
When it comes to geographical distribution, 59.4% of AI stakeholders are based in the provinces of Liège and Walloon Brabant and, to a lesser extent, in Hainaut. Overall distribution is very similar to that observed for the digital sector as a whole.
AI and specialist fields in Wallonia
Unlike, for example, the IoT ecosystem, AI stakeholders are not necessarily targeting specific sectors. Logically, while just 9 companies say that they are mainly targeting the B2C market, 60.7% of stakeholders are not targeting any particular sector. This is probably due to the impact of the fact that AI is, by its nature, a “multi-application” field.
Stakeholders identified as being interested in specific markets:
- Health (29 businesses)
- Aviation / Space (13 businesses)
- Industry (13 businesses)
- Transport and Logistics (9 businesses)
- Finance (8 businesses)
- Automotive (8 businesses)
- Creative Industries (7 businesses)
- Distribution (6 businesses)
The sectors targeted by the remaining 33 specialist companies are evenly distributed between agriculture, defence, government, energy, etc.
Overall, while private provision in terms of AI is still fairly limited in Wallonia, we can see that there is real potential when it comes to offering advanced technological solutions aimed at businesses (both in Wallonia and internationally).
A rich, diverse range of AI services
AI provision in Wallonia is mainly made up of digital services and software development (364 of the total number of different solutions were in this category). However, when a company offers multiple products and services, AI is strongly connected to other advanced technologies used or developed by digital companies in Wallonia. This demonstrates a willingness to develop solutions that:
- complement other advanced technologies (AI, blockchain, robotics etc.),
- are integrated into state-of-the-art products, services, platforms, processes or networks.
So 25.6% of Walloon companies involved in AI offer services related to the IoT, 11.5% robotics, 10.5% AR/VR and 7.3% digital twins and simulation. Of course, many companies also offer data analytics or machine learning services, due to the fact that these technologies are closely interconnected with AI.
These “Advanced” solutions connected to AI can be broken down as follows:
- IoT (49)
- Robotics (22)
- AR/VR (20)
- Digital twins / simulation (14)
- Blockchain (9)
- HPC (7)
- Drones (5)
- 3D printing (4)
- Nanotechnologies (1)
- Data Analytics, ML, Cloud (75)
AI : international opportunities
Although 50 AI companies are members of the AWEX network, Wallonia’s AI ecosystem is not (yet?) particularly geared towards the international market. 24 stakeholders (12.6%) involved in AI have taken part in international missions and only 3 have taken part in more than 10 missions.
This can of course be attributed in part to the fact that this sector is so young. However, it seems to be crucial for a company involved in AI to establish itself in international markets quickly if it wants to guarantee its growth and pursue its commitment to innovation, because the regional and even national markets are currently so narrow. Indeed, international missions would give them the chance to:
- Build up the expertise they need to develop their business while they are abroad;
- Seek out new clients, expand their network and target new markets;
- Develop international partnerships with economic and research stakeholders.
AI and other ecosystems
Walloon companies offering AI solutions are involved in a large number of specialist or emerging digital ecosystems showcased by Digital Wallonia.
For example, 46 companies offer AI solutions for the Industry 4.0 ecosystem. This cutting-edge technology represents more than a quarter of all digital businesses offering products and services designed to transform Walloon industry. The other ecosystems are Big & Smart Data, e-health and Smart Cities.
More specifically :
- Industry 4.0: 46 businesses
- Big & Smart Data: 44 businesses
- e-Health: 35 businesses
- Smart Cities: 30 businesses
- Cybersecurity: 16 businesses
- AdTech: 13 businesses
- Energy: 12 businesses
- EdTech: 10 businesses
- Smart Farming: 9 businesses
- Coding: 9 businesses
- MedTech: 7 businesses
- Smart Grid: 7 businesses
- Smart Logistics: 7 businesses
- Other ecosystems: 44 businesses
AI : real momentum in this region
Artificial Intelligence is undeniably the flagship digital technology, which will preside over the economic development of future economies. The opportunities and fields of application seem almost unlimited. Wallonia boasts significant advantages in this area, particularly when it comes to research. This is mainly concentrated in universities (almost 600 researchers are totally or partially involved in this area), and covers a range of fields including voice and image recognition, decision making, market research etc.
A growing number of startups and more mature companies are offering or developing AI solutions with clear economic potential, especially in fields of application such as Industry 4.0, e-health, Smart Cities etc.
However, with the exception of a few well-known leaders, it is taking a long time for these research capabilities and Proofs of concept to be translated into commercial and international success. More than ever before, it is worth insisting on the need for cooperation between research centres and businesses (especially startups) in order to make the most of Wallonia’s expertise in this field as soon as possible, just as we need to encourage our AI providers to embrace international markets, if necessary by taking targeted measures. Training opportunities also need to be developed, both at universities and in terms of professional training. Some initiatives have recently been launched in this area by universities, skills centres and the Forem, and these need to be consolidated.
Alongside this, we need to offer support to businesses outside the digital sector when it comes to adopting AI with tailor-made programmes targeting their core business. AI is still overlooked and even rejected by potential buyers for a range of reasons, as demonstrated by a 2016-2017 Forrester study on the obstacles to adoption:
- There is no defined business case: 42%
- Not clear what AI can be used for: 39%
- Don’t have the required skills: 33%
- Need first to invest in modernising our data management platform: 29%
- Don't have the budget: 23%
- Not certain what is needed for implementing an AI system: 19%
- AI systems are not proven: 14%
- Do not have the right processes or governance: 13%
- AI is a lot of hype with little substance: 11%
In order to encourage the development of innovative solutions and support adoption, the DigitalWallonia4AI programme can and should become a major tool of the Region's economic development strategy, in partnership with initiatives such as the Trail Institute and the AI Network.
The analysis was carried out on all the data from the Digital Wallonia platform relating to stakeholders in Wallonia involved in supply, demand and support associated with artificial intelligence technology. This data guarantees a true representation of Wallonia’s situation. The status report was completed in November 2020.
By definition, these results are not static. They are evolving according to how businesses are developing and how economic trends are moving and on the basis of updates provided by the Agence du Numérique and its partners.