AI excellence: thriving from the lab to the market

Strategies for developing and adopting trustworthy AI

The strategic framework set out in the Communication on boosting startups and innovation in trustworthy artificial intelligence aims to foster an innovative, fair, open and contestable AI market. It not only bolstering European companies at home but also empowers them to compete confidently on the global stage. This Communication build upon the existing European approach to excellence in AI, in particular the Coordinated Plan on AI.

The 2021 review of the Coordinated Plan includes proposals to make the EU the place where artificial intelligence (AI) excellence thrives from the lab to the market by promoting research in AI, encouraging its uptake, and funding innovation. The Commission strategy for supporting the development and adoption of trustworthy AI solutions is a values-driven approach which covers the whole lifecycle and focuses on building an ecosystem of excellence. The key actions seek to:

• foster collaboration between stakeholders through Public-Private Partnerships (PPP)
• build and mobilise AI research capabilities
• provide facilities for developers and companies to test and experiment with their latest AI-based technologies in real-world environments
• develop a European marketplace for trustworthy AI solutions, connecting resources and services
• fund and scale innovative ideas and solutions for AI

European partnership on AI, Data and Robotics

The European Partnership on AI, Data and Robotics Association (ADRA) was officially launched in June 2021 with the signature of the Memorandum of Understanding between the European Commission and the ADR association. ADRA was awarded 2.6 Billion Euros under the funding programmes Horizon 2020 (2014-2020) and Horizon Europe (2021-2027). This public-private partnership brings together communities from three key communities – AI, Data and Robotics -  and five organisations: BDVA, CLAIRE, ELLIS, EurAI and euRobotics .

In 2024, ADRA counts more than 133 members, from startups, SMEs, big industry as well as research organisations and RTOs. In November 2023, ADRA organised its first event - the ADR Forum which focused on Generative AI. ADRA engages with many stakeholders (including the industry), to tackle the challenges of generative AI. This lead to the establishment  of 2 taskforces in 2023:

• GenAI - dedicated to addressing challenges related to generative AI.
• Generative AI for Robotics - focused on generative AI applications in the field of robotics.

World-class testing and experimentation facilities

From January 2023, 4 sectorial testing and experimentation facilities (TEFs) have started providing European SMEs with the means to test their latest AI-based technologies  (Technology Readiness Levels 6-8)  in real or close-to-real conditions. TEF services involve support for full integration, the testing, experimentation, and validation of AI-based solutions in the sectors served by the four TEFs:

The TEFs are co-funded by the Commission and Member States with a total investment of over 220 million Euros and help innovators to test their close-to-market AI solutions before introducing them to the market.

In March 2024, the four Testing and Experimentation Facilities (TEFs) held their official event, signalling their readiness to engage with businesses. These TEFs provide technical assistance to AI innovators, allowing them to evaluate and experiment with their cutting-edge AI software and hardware in real-world scenarios (specifically, at Technology Readiness Levels 6-8). The services offered encompass full integration support, as well as testing, experimentation, and validation of AI-based solutions across the sectors served by these four TEFs:

• Healthcare: TEF-Health
• Agri-Food: agrifoodTEF
• Manufacturing: AI-MATTERS
• Smart Cities & Communities: Citcom.AI

The European AI-on-demand platform

The AI on demand platform aims at building a bridge between European AI research, industry and public services. It is a catalyst to light up collaboration between academia and businesses but also among other stakeholders, reducing the gap between AI research and innovators. The European AI-on-Demand wants to enable the latest  AI advances to be rapidly transformed into real AI products and services deployed into the market.

The platform is supported by an ecosystem of projects funded under Horizon 2020 and Horizon Europe (focus on research), and Digital Europe programme (focus on industry and public administration). The current version of the platform includes several services:

• An assets catalogue
• A powerful Virtual AI lab where researchers can explore the contents related to AIoD (Artificial Intelligence of Data) across multiple platforms
• The ability to create their own libraries
• Development, training, and sharing of AI pipelines
• An extended and adapted marketplace catering to industry needs in upcoming versions

European Digital Innovation Hubs

A network of more than 200 European Digital Innovation Hubs (EDIHs) covering all regions of Europe aims to stimulate the broad uptake of AI, HPC, cybersecurity and other digital technologies. EDIHs help all companies seeking to use AI technologies to become more competitive on business/production processes, products or services. As ‘one-stop-shops’, EDIHs can promote locally the use of tools from the AI-on-demand platform to industry, including small and medium-sized enterprises, and the public sector. They will assist companies to innovate their new products and services with AI and stimulate adoption by helping to make them market ready.

At present, there are around 150 EDIHs speciliased in AI and decision support. Several trainings and webinars on the topic of AI (including generative AI), are frequently organised. Also, a dedicated working group on AI in public administration was created.

European Digital Infrastructure Consortium

The European Digital Infrastructure Consortium (EDIC) is a legal framework designed to help Member States collaboratively establish and execute multi-country projects. Together with the Commission, Member States are investing EUR 100 million in two AI-related EDICs:

• The Alliance for Language Technologies (ALT-EDIC) focuses on developing large language models for European regional and official languages. The goal is to create a common European infrastructure in language technologies, which will enhance digital competitiveness, preserve linguistic diversity, and promote cultural richness in Europe
• The Networked Local Digital Twins towards the CitiVERSE (LDT CitiVERSE EDIC) aims to connect existing local digital twins across Europe. Local digital twins are virtual representations of a city’s physical assets, processes, and systems. By integrating these digital twins, the project seeks to create real-time simulation models that can improve urban planning and management.