Artificial intelligence is no longer a matter of speculation. For Luxembourg businesses, it is already an economic reality, and indeed a test of competitiveness. During a joint presentation on stage at Nexus Luxembourg, representatives from PwC and Microsoft, 
Its most strategic sectors (finance, professional services, IT) are also among those most exposed to AI. Behind the promises of automation, the stakes go far beyond mere time savings. It touches on productivity, revenue models, data quality, skills and organisations’ ability to move from experimentation to industrialisation. Because not everyone will capture value at the same pace. Five figures help to gauge the scale of this shift.
50%
This is the share of Luxembourg’s GDP accounted for by finance, professional services and IT, which are also the sectors most exposed to AI. This proves, if proof were needed, that AI is not a peripheral issue: the sectors most exposed – finance, professional services and IT – account for around half of GDP.
“This means that Luxembourg is likely at a turning point. Our economy is heavily exposed to AI. Several decades ago, Luxembourg made certain choices. It identified opportunities for growth and accelerated the adoption of technologies that gave it a competitive advantage. Today, Luxembourg has one of the highest per capita GDPs in the world. But there is a problem. In recent years, GDP growth has stalled and is even lower than that of the eurozone. That is the problem. Luxembourg, its sectors and its organisations are therefore at a new crossroads. This is an opportunity to regain a competitive edge. But it is also a risk for organisations based in Luxembourg: that of failing to adopt AI, or failing to adopt it quickly enough, in a globalised economy,” said Patrice Witz.
5–12% increase in productivity
“The OECD published a report in June 2025 showing productivity gains by sector. The sectors most affected are as follows: the financial sector, with productivity gains of 5 to 12 per cent, IT services and professional services,” added Patrice Witz.
According to Pierre-Adrien Grange, “when we look at organisations today, one thing is clear: those currently winning the AI race are not just the ones experimenting with AI. They are the ones that manage to roll it out across the entire organisation.” He identifies four key pillars.
Firstly, they aim to enhance the employee experience. They use AI to empower every individual, regardless of their role or position, in order to eliminate tedious tasks and allow employees to focus on what really matters: human judgement, empathy and creativity. Secondly, they are reinventing customer engagement, making it smarter, more real-time and, crucially, more personalised and cost-effective. The third pillar involves reshaping the operational model and business processes with AI. This is not simply about automation, but about truly rethinking the entire operational model of the business by integrating it with AI. By doing these three things, organisations can unlock something very powerful: the fourth pillar, which involves shifting the innovation curve.
But first and foremost, one essential requirement was highlighted: data quality. “But there is one thing, one indispensable foundation for enabling this transformation: having the right database, the right governance and, above all, trust. When we talk here of ‘intelligence + trust’ as the foundation of our framework, it means that an organisation must strike the right balance between integrating AI everywhere, across all its business processes, and ensuring that this AI is reliable, explainable and does not undermine trust. It must also be aligned with regulatory obligations and expectations. It is really when we combine these two dimensions that we enable real transformation: not just greater efficiency, but also an increased ability to help employees innovate much more quickly.”
20–40% increase in efficiency
“First and foremost, when we talk about AI, we tend to think mainly of efficiency gains: the automation of repetitive tasks, and a reduction in manual work in back-office functions or certain activities. The most advanced organisations are already achieving efficiency gains of 20 to 40 per cent in certain specific areas. But that is what we call the defensive approach,” said the PwC partner.
But there is another approach: the offensive one. “Today, 30% of financial institutions worldwide are actively planning to work on new revenue models based on AI: new products, new services, new value propositions for customers, and the development of business activities made possible by AI that were previously structurally impossible to achieve. This is a topic that must be at the heart of strategic discussions,” he adds.
“In our latest Work Trend Index report, eight out of ten professionals working in advanced organisations – that is, organisations that have truly embraced AI – say they are now producing work that would have been simply impossible a year ago. That is what it means to bend the innovation curve. It shows that this transformation is already evident in the data and in companies’ business results,” added the country lead for Microsoft Luxembourg.
74%
According to PwC, 74% of the value generated by AI is captured by just 20% of organisations. In other words, the risk is not that we will ‘miss out on AI’ in general, but that a minority will capture the lion’s share of the value.
“I think it’s important for you to understand this: for an organisation, the race for AI isn’t about choosing the best models. There is a lot of talk, particularly in the news, about the emergence of all these models – Mistral, OpenAI, Anthropic – and all the other players who are releasing new models at a very rapid pace. But the real added value, the real differentiator for a company, is having the platform capable of orchestrating these models,” added Pierre-Adrien Grange.
+15
In terms of GDP percentage points, this is what the adoption of AI could generate. In other words, AI is not just a productivity tool, but a macroeconomic driver. “The adoption of AI could increase global GDP by a further 15 percentage points over the next decade. This is a significant impact on GDP growth, not only globally but also within the context of the global economy. Which economies will pull ahead? Which ones will fall behind? And by how much? That remains to be seen,” emphasised the PwC partner.
