Artificial intelligence is great… but for what purpose? Identifying concrete use cases is a prerequisite before even considering the implementation of these technologies.
Jean-Michel Gaudron
Artificial intelligence (AI) in business is becoming an increasingly tangible reality. Whether it is used to optimize internal processes, improve customer relations, innovate products and services, or strengthen risk management and compliance, AI tools offer an impressive range of possibilities — provided they are used correctly.
The recent national survey conducted by Luxinnovation and FEDIL, in collaboration with the Luxembourg Digital Innovation Hub (L-DIH), shows that a large majority of surveyed companies (over 70%) see the adoption of these technologies as an opportunity to optimise processes and controls and to reduce costs. Around 60% also view AI as a driver of quality improvement and a valuable tool for better decision-making.
The survey reveals a notable trend in corporate adoption of AI: 63% of respondents report being at advanced stages of AI maturity, with applications ranging from targeted production use cases to company-wide implementations.
More specifically, 27% of respondents say they already have AI applications in development or production, while 23% are in the experimentation and functional model phase.
63% of respondents report being at advanced stages of AI maturity
A significant number of companies have therefore already integrated AI at the enterprise level, reflecting a strong commitment to embedding these technologies into their operations.
However, more than a third of companies (37%) are still in the early stages of AI adoption, with a relatively low level of maturity. These companies are mainly focused on awareness initiatives, with 16% currently engaged in such efforts. This trend is even more pronounced in the manufacturing sector, where the figure reaches 23%.
“It is important to explore how these companies plan to transition from awareness to practical AI adoption — including the studies they conduct, the use cases they identify, and the support they require during this transition,” the study authors emphasize.
The main reasons cited by the 30% of respondents who have not adopted generative AI (GenAI) are the prioritization of other strategic projects and the lack of clearly defined and relevant use cases. Concerns about data preparation and security/privacy are also major factors. The study further highlights that a lack of awareness or understanding of GenAI’s potential is an additional barrier for some.
“This underscores the continued need for training, the sharing of successful case studies, and robust security frameworks to encourage broader adoption,” the study notes.
For many companies with lower levels of AI maturity, the lack of internal expertise remains a major obstacle to adoption. This is a challenge further compounded by the absence of clearly defined use cases.
It is crucial to choose the right technologies and to take into account the legal and regulatory implications. Philippe Mayer, Luxinnovation
“In this context, a service such as ‘Test Before Invest’ can be extremely useful to guide companies in their technology choices,” explains Mickael Desloges, Senior Advisor at Luxinnovation.
Offered as part of the Luxembourg Digital Innovation Hub in partnership with the Luxembourg Institute of Science and Technology (LIST), this service gives companies the opportunity to validate AI use cases with their own data — without the need to make an immediate, long-term commitment.
The Fit 4 AI programme, also offered by Luxinnovation, follows the same logic of helping companies find the best way to integrate artificial intelligence into their business processes.
Based on a diagnostic and preliminary assessment carried out by an experienced consultant, the programme makes it possible to define — in close consultation with the company — its key challenges, objectives, and priority development areas.
Despite its democratisation, AI requires expertise more than ever to define use cases. Philippe Mayer, Luxinnovation
This study is financially supported by state aid, granted in the form of a capital subsidy by the Ministry of the Economy.
“Despite its democratisation, AI requires expertise more than ever to define use cases and transform them into concrete projects, estimate the potential benefits for the company, and perform cost-benefit analyses. It is crucial to choose the right technologies and to take into account the legal and regulatory implications,” explains Philippe Mayer, Deputy CEO of Luxinnovation.
This qualitative and quantitative analysis of the available internal data allows the company to clearly determine its current situation, set its future goals, and measure its digital maturity.
It can then more easily identify relevant potential use cases based on proposed technology choices, budget considerations, risk assessments, and ROI forecasts.
Finally, a roadmap is drawn up for the project’s implementation, including a timeline and action plan, quantified needs in terms of internal and external resources, and equipment requirements.
The process can also directly lead to referrals to suitable training programmes, support schemes, or partner organisations.
